Melanoma

When clinicians in a large-scale study viewed a series of digital images that showed skin diseases across skin tones and were asked to make a diagnosis, the accuracy was 38% among dermatologists and 19% among primary care physicians (PCPs). But when decision support from a deep learning system (DLS) was introduced, diagnostic accuracy increased by 33% among dermatologists and 69% among PCPs, results from a multicenter study showed.

However, the researchers found that across all images, diseases in dark skin (Fitzpatrick skin types 5 and 6) were diagnosed less accurately than diseases in light skin (Fitzpatrick skin types 1-4).

“These results contribute to an emerging literature on diagnostic accuracy disparities across patient skin tones and present evidence that the diagnostic accuracy of medical professionals on images of dark skin is lower than on images of light skin,” researchers led by Matthew Groh, PhD, of Northwestern University’s Kellogg School of Management, wrote in their study, published online in Nature Medicine.

For the study, 389 board-certified dermatologists and 450 PCPs in 39 countries were presented with 364 images to view spanning 46 skin diseases and asked to submit up to four differential diagnoses. Nearly 80% of the images were of 8 diseases: atopic dermatitis, cutaneous T-cell lymphoma (CTCL), dermatomyositis, lichen planus, Lyme disease, pityriasis rosea, pityriasis rubra pilaris, and secondary syphilis.

Dermatologists and PCPs achieved a diagnostic accuracy of 38% and 19%, respectively, but both groups of clinicians were 4 percentage points less accurate for diagnosis of images of dark skin as compared with light skin. With assistance from DLS decision support, diagnostic accuracy increased by 33% among dermatologists and 69% among primary care physicians. Among dermatologists, DLS support generally increased diagnostic accuracy evenly across skin tones. However, among PCPs, DLS support increased their diagnostic accuracy more in light skin tones than in dark ones.

In the survey component of the study, when the participants were asked, “Do you feel you received sufficient training for diagnosing skin diseases in patients with skin of color (non-white patients)?” 67% of all PCPs and 33% of all dermatologists responded no. “Furthermore, we have found differences in how often BCDs [board-certified dermatologists] and PCPs refer patients with light and dark skin for biopsy,” the authors wrote. “Specifically, for CTCL (a life-threatening disease), we found that both BCDs and PCPs report that they would refer patients for biopsy significantly more often in light skin than dark skin. Moreover, for the common skin diseases atopic dermatitis and pityriasis rosea, we found that BCDs report they would refer patients for biopsy more often in dark skin than light skin, which creates an unnecessary overburden on patients with dark skin.”

In a press release about the study, Dr. Groh emphasized that he and other scientists who investigate human-computer interaction “have to find a way to incorporate underrepresented demographics in our research. That way we will be ready to accurately implement these models in the real world and build AI systems that serve as tools that are designed to avoid the kind of systematic errors we know humans and machines are prone to. Then you can update curricula, you can change norms in different fields and hopefully everyone gets better.”

When clinicians in a large-scale study viewed a series of digital images that showed skin diseases across skin tones and were asked to make a diagnosis, the accuracy was 38% among dermatologists and 19% among primary care physicians (PCPs). But when decision support from a deep learning system (DLS) was introduced, diagnostic accuracy increased by 33% among dermatologists and 69% among PCPs, results from a multicenter study showed.

However, the researchers found that across all images, diseases in dark skin (Fitzpatrick skin types 5 and 6) were diagnosed less accurately than diseases in light skin (Fitzpatrick skin types 1-4).

“These results contribute to an emerging literature on diagnostic accuracy disparities across patient skin tones and present evidence that the diagnostic accuracy of medical professionals on images of dark skin is lower than on images of light skin,” researchers led by Matthew Groh, PhD, of Northwestern University’s Kellogg School of Management, wrote in their study, published online in Nature Medicine.

For the study, 389 board-certified dermatologists and 450 PCPs in 39 countries were presented with 364 images to view spanning 46 skin diseases and asked to submit up to four differential diagnoses. Nearly 80% of the images were of 8 diseases: atopic dermatitis, cutaneous T-cell lymphoma (CTCL), dermatomyositis, lichen planus, Lyme disease, pityriasis rosea, pityriasis rubra pilaris, and secondary syphilis.

Dermatologists and PCPs achieved a diagnostic accuracy of 38% and 19%, respectively, but both groups of clinicians were 4 percentage points less accurate for diagnosis of images of dark skin as compared with light skin. With assistance from DLS decision support, diagnostic accuracy increased by 33% among dermatologists and 69% among primary care physicians. Among dermatologists, DLS support generally increased diagnostic accuracy evenly across skin tones. However, among PCPs, DLS support increased their diagnostic accuracy more in light skin tones than in dark ones.

In the survey component of the study, when the participants were asked, “Do you feel you received sufficient training for diagnosing skin diseases in patients with skin of color (non-white patients)?” 67% of all PCPs and 33% of all dermatologists responded no. “Furthermore, we have found differences in how often BCDs [board-certified dermatologists] and PCPs refer patients with light and dark skin for biopsy,” the authors wrote. “Specifically, for CTCL (a life-threatening disease), we found that both BCDs and PCPs report that they would refer patients for biopsy significantly more often in light skin than dark skin. Moreover, for the common skin diseases atopic dermatitis and pityriasis rosea, we found that BCDs report they would refer patients for biopsy more often in dark skin than light skin, which creates an unnecessary overburden on patients with dark skin.”

In a press release about the study, Dr. Groh emphasized that he and other scientists who investigate human-computer interaction “have to find a way to incorporate underrepresented demographics in our research. That way we will be ready to accurately implement these models in the real world and build AI systems that serve as tools that are designed to avoid the kind of systematic errors we know humans and machines are prone to. Then you can update curricula, you can change norms in different fields and hopefully everyone gets better.”

When clinicians in a large-scale study viewed a series of digital images that showed skin diseases across skin tones and were asked to make a diagnosis, the accuracy was 38% among dermatologists and 19% among primary care physicians (PCPs). But when decision support from a deep learning system (DLS) was introduced, diagnostic accuracy increased by 33% among dermatologists and 69% among PCPs, results from a multicenter study showed.

However, the researchers found that across all images, diseases in dark skin (Fitzpatrick skin types 5 and 6) were diagnosed less accurately than diseases in light skin (Fitzpatrick skin types 1-4).

“These results contribute to an emerging literature on diagnostic accuracy disparities across patient skin tones and present evidence that the diagnostic accuracy of medical professionals on images of dark skin is lower than on images of light skin,” researchers led by Matthew Groh, PhD, of Northwestern University’s Kellogg School of Management, wrote in their study, published online in Nature Medicine.

For the study, 389 board-certified dermatologists and 450 PCPs in 39 countries were presented with 364 images to view spanning 46 skin diseases and asked to submit up to four differential diagnoses. Nearly 80% of the images were of 8 diseases: atopic dermatitis, cutaneous T-cell lymphoma (CTCL), dermatomyositis, lichen planus, Lyme disease, pityriasis rosea, pityriasis rubra pilaris, and secondary syphilis.

Dermatologists and PCPs achieved a diagnostic accuracy of 38% and 19%, respectively, but both groups of clinicians were 4 percentage points less accurate for diagnosis of images of dark skin as compared with light skin. With assistance from DLS decision support, diagnostic accuracy increased by 33% among dermatologists and 69% among primary care physicians. Among dermatologists, DLS support generally increased diagnostic accuracy evenly across skin tones. However, among PCPs, DLS support increased their diagnostic accuracy more in light skin tones than in dark ones.

In the survey component of the study, when the participants were asked, “Do you feel you received sufficient training for diagnosing skin diseases in patients with skin of color (non-white patients)?” 67% of all PCPs and 33% of all dermatologists responded no. “Furthermore, we have found differences in how often BCDs [board-certified dermatologists] and PCPs refer patients with light and dark skin for biopsy,” the authors wrote. “Specifically, for CTCL (a life-threatening disease), we found that both BCDs and PCPs report that they would refer patients for biopsy significantly more often in light skin than dark skin. Moreover, for the common skin diseases atopic dermatitis and pityriasis rosea, we found that BCDs report they would refer patients for biopsy more often in dark skin than light skin, which creates an unnecessary overburden on patients with dark skin.”

In a press release about the study, Dr. Groh emphasized that he and other scientists who investigate human-computer interaction “have to find a way to incorporate underrepresented demographics in our research. That way we will be ready to accurately implement these models in the real world and build AI systems that serve as tools that are designed to avoid the kind of systematic errors we know humans and machines are prone to. Then you can update curricula, you can change norms in different fields and hopefully everyone gets better.”

SAN DIEGO — Every year, about 4000 patients in the United States are diagnosed with periocular melanoma, according to Geva Mannor, MD, MPH.

“Though rare, the incidence is thought to be slightly increasing, while the onset tends to occur in patients over age 40 years,” Dr. Mannor, an oculofacial plastic surgeon in the division of ophthalmology at Scripps Clinic, San Diego, said at the annual Cutaneous Malignancy Update. “It may be more common in males and welding is a risk factor. Pain and vision loss are late symptoms, and there are amelanotic variants. Gene expression profiling and other genetic testing can predict metastasis, especially expression of BRCA1-associated protein 1 (BAP1).”

An estimated 83% of periocular melanoma cases are choroid (which involve the intraocular part of the eye and the uvea), about 10% involve the eyelid, while about 1%-3% involve the conjunctiva. Extrapolating from the best available data, Dr. Mannor said that the annual incidence of choroidal melanoma in the United States is less than 2500, the annual incidence of eyelid melanoma is less than 750, and the annual incidence of conjunctival melanoma is less than 250 — much lower than for cutaneous melanomas. Put another way, the ratio between cutaneous and choroid melanoma is 80:1, the ratio between cutaneous and eyelid melanoma is 266:1, and the ratio between cutaneous and conjunctival melanoma is 800:1.

Dr. Mannor

According to an article published in 2021 on the topic, risk factors for periocular melanoma include light eye color (blue/gray; relative risk, 1.75), fair skin (RR, 1.80), and inability to tan (RR, 1.64), but not blonde hair. A review of 210 patients with melanoma of the eyelid from 11 studies showed that 57% were located on the lower lid, 13% were on the upper lid (“I think because the brow protects sun exposure to the upper lid,” Dr. Mannor said), 12% were on the brow, 10% were on the lateral canthus, and 2% were on the medial canthus. In addition, 35% of the eyelid melanomas were superficial spreading cases, 31% were lentigo maligna, and 19% were nodular. The mean Breslow depth was 1.36 mm and the mortality rate was 4.9%.

Dr. Mannor said that cheek and brow melanomas can extend to the inside of the eyelid and conjunctiva. “Therefore, you want to examine the inside of upper and lower eyelids,” he said at the meeting, which was hosted by the Scripps Cancer Center. “Margin-control excision of lid melanoma is the standard treatment.”

On a related note, he said that glaucoma eye drops that contain prostaglandin F2alpha induce cutaneous and iris pigmentation with varying rates depending on the specific type of prostaglandin. “There have been case reports of these eye drops causing periocular pigmentation that masquerades as suspicious, melanoma-like skin cancer, often necessitating a skin biopsy,” he said.

Dr. Mannor reported having no disclosures.

SAN DIEGO — Every year, about 4000 patients in the United States are diagnosed with periocular melanoma, according to Geva Mannor, MD, MPH.

“Though rare, the incidence is thought to be slightly increasing, while the onset tends to occur in patients over age 40 years,” Dr. Mannor, an oculofacial plastic surgeon in the division of ophthalmology at Scripps Clinic, San Diego, said at the annual Cutaneous Malignancy Update. “It may be more common in males and welding is a risk factor. Pain and vision loss are late symptoms, and there are amelanotic variants. Gene expression profiling and other genetic testing can predict metastasis, especially expression of BRCA1-associated protein 1 (BAP1).”

An estimated 83% of periocular melanoma cases are choroid (which involve the intraocular part of the eye and the uvea), about 10% involve the eyelid, while about 1%-3% involve the conjunctiva. Extrapolating from the best available data, Dr. Mannor said that the annual incidence of choroidal melanoma in the United States is less than 2500, the annual incidence of eyelid melanoma is less than 750, and the annual incidence of conjunctival melanoma is less than 250 — much lower than for cutaneous melanomas. Put another way, the ratio between cutaneous and choroid melanoma is 80:1, the ratio between cutaneous and eyelid melanoma is 266:1, and the ratio between cutaneous and conjunctival melanoma is 800:1.

Dr. Mannor

According to an article published in 2021 on the topic, risk factors for periocular melanoma include light eye color (blue/gray; relative risk, 1.75), fair skin (RR, 1.80), and inability to tan (RR, 1.64), but not blonde hair. A review of 210 patients with melanoma of the eyelid from 11 studies showed that 57% were located on the lower lid, 13% were on the upper lid (“I think because the brow protects sun exposure to the upper lid,” Dr. Mannor said), 12% were on the brow, 10% were on the lateral canthus, and 2% were on the medial canthus. In addition, 35% of the eyelid melanomas were superficial spreading cases, 31% were lentigo maligna, and 19% were nodular. The mean Breslow depth was 1.36 mm and the mortality rate was 4.9%.

Dr. Mannor said that cheek and brow melanomas can extend to the inside of the eyelid and conjunctiva. “Therefore, you want to examine the inside of upper and lower eyelids,” he said at the meeting, which was hosted by the Scripps Cancer Center. “Margin-control excision of lid melanoma is the standard treatment.”

On a related note, he said that glaucoma eye drops that contain prostaglandin F2alpha induce cutaneous and iris pigmentation with varying rates depending on the specific type of prostaglandin. “There have been case reports of these eye drops causing periocular pigmentation that masquerades as suspicious, melanoma-like skin cancer, often necessitating a skin biopsy,” he said.

Dr. Mannor reported having no disclosures.

SAN DIEGO — Every year, about 4000 patients in the United States are diagnosed with periocular melanoma, according to Geva Mannor, MD, MPH.

“Though rare, the incidence is thought to be slightly increasing, while the onset tends to occur in patients over age 40 years,” Dr. Mannor, an oculofacial plastic surgeon in the division of ophthalmology at Scripps Clinic, San Diego, said at the annual Cutaneous Malignancy Update. “It may be more common in males and welding is a risk factor. Pain and vision loss are late symptoms, and there are amelanotic variants. Gene expression profiling and other genetic testing can predict metastasis, especially expression of BRCA1-associated protein 1 (BAP1).”

An estimated 83% of periocular melanoma cases are choroid (which involve the intraocular part of the eye and the uvea), about 10% involve the eyelid, while about 1%-3% involve the conjunctiva. Extrapolating from the best available data, Dr. Mannor said that the annual incidence of choroidal melanoma in the United States is less than 2500, the annual incidence of eyelid melanoma is less than 750, and the annual incidence of conjunctival melanoma is less than 250 — much lower than for cutaneous melanomas. Put another way, the ratio between cutaneous and choroid melanoma is 80:1, the ratio between cutaneous and eyelid melanoma is 266:1, and the ratio between cutaneous and conjunctival melanoma is 800:1.

Dr. Mannor

According to an article published in 2021 on the topic, risk factors for periocular melanoma include light eye color (blue/gray; relative risk, 1.75), fair skin (RR, 1.80), and inability to tan (RR, 1.64), but not blonde hair. A review of 210 patients with melanoma of the eyelid from 11 studies showed that 57% were located on the lower lid, 13% were on the upper lid (“I think because the brow protects sun exposure to the upper lid,” Dr. Mannor said), 12% were on the brow, 10% were on the lateral canthus, and 2% were on the medial canthus. In addition, 35% of the eyelid melanomas were superficial spreading cases, 31% were lentigo maligna, and 19% were nodular. The mean Breslow depth was 1.36 mm and the mortality rate was 4.9%.

Dr. Mannor said that cheek and brow melanomas can extend to the inside of the eyelid and conjunctiva. “Therefore, you want to examine the inside of upper and lower eyelids,” he said at the meeting, which was hosted by the Scripps Cancer Center. “Margin-control excision of lid melanoma is the standard treatment.”

On a related note, he said that glaucoma eye drops that contain prostaglandin F2alpha induce cutaneous and iris pigmentation with varying rates depending on the specific type of prostaglandin. “There have been case reports of these eye drops causing periocular pigmentation that masquerades as suspicious, melanoma-like skin cancer, often necessitating a skin biopsy,” he said.

Dr. Mannor reported having no disclosures.

SAN DIEGO — According to data from the American Cancer Society (ACS), cutaneous melanoma was the fifth most common cancer in 2023, with an estimated 97,610 new cases and 7,990 deaths, following cancer of the colorectal area, lung and bronchus, prostate, and breast.

“The incidence of melanoma seems to have continued to go up since the early 1990s,” David E. Kent, MD, a dermatologist at Skin Care Physicians of Georgia, Macon, said at the annual Cutaneous Malignancy Update. “The death rates have been flat and may have slightly decreased.”

In 2024, the ACS estimates that about 100,640 new melanomas will be diagnosed in the United States (59,170 in men and 41,470 in women), and about 8,290 people are expected to die of melanoma (5,430 men and 2,860 women). Meanwhile, the lifetime risk of melanoma is about 3% (1 in 33) for Whites, 0.1% (1 in 1,000) for Blacks, and 0.5% (1 in 200) for Hispanics. In 2019, there were an estimated 1.4 million people in the United States living with cutaneous melanoma, and the overall 5-year survival is 93.7%.

Epidemiologic studies show an increase in melanoma incidence, primarily among White populations. “This is believed to be due to sun exposure, changes in recreational behaviors, and tanning bed exposures,” said Dr. Kent, who holds a faculty position in the department of dermatology at the Medical College of Georgia, Augusta. Increased surveillance and diagnosis also play a role. In the medical literature, annual increases in melanoma incidence vary from 3%-7% per year, “which means that the rate is doubling every 10-20 years,” he said, noting that annual melanoma costs are approximately $3.3 billion.

While incidence rates are lower in non-White, non-Hispanic populations, poor outcomes are disproportionately higher in persons of color, according to a 2023 paper. Black individuals present at diagnosis with more advanced stage disease and are 1.5 times more likely to die from melanoma, he said, while Hispanics are 2.4 times more likely to present with stage III disease and 3.6 times more likely to have distant metastases. Persons of color also have higher rates of mucosal, acral lentiginous, and subungual melanoma.

Risk Factors

Known genetic risk factors for melanoma include having skin types I and II, particularly those with light hair, light eyes, and freckling, and those with a family history have a twofold increased risk. Also, up to 40% of genetic cases are from inherited mutations in CDKN2A, CDK4, BAP1 and MCR1. Other genetic-related risk factors include the number and size of nevi, having atypical nevus syndrome, DNA repair defects, large congenital nevi, and a personal history of melanoma.

Dr. Kent said that genetic testing for melanoma is warranted for individuals who meet criteria for the “rule of 3s.” He defined this as three primary melanomas in an individual, or three cases of melanoma in first- or second-degree relatives, or two cases of melanoma and one pancreatic cancer or astrocytoma in first or second-degree relatives, or one case of melanoma and two of pancreatic cancer/astrocytoma in first- or second-degree relatives.

The main environmental risk factor for melanoma is exposure to ultraviolet radiation. Individuals can track their UV exposure with a variety of wearable devices and apps, including SunSense One Digital UV Tracker, the SunSense App, the UV Index Widget, the SunSmart Global UV App, the SunKnown UV light photometer, and the EPA’s UV Index Mobile App. Other environment-related risk factors include having a high socioeconomic status (SES), being immunosuppressed, as well as exposure to heavy metals, insecticides, or hormones; and ones distance from the equator.

In a study published in 2023, researchers investigated the risk factors associated with first and second primary melanomas in 38,845 patients who were followed in Australia between 2011 and 2018. During a median follow-up of 7.4 years, 1212 patients (3.1%) had a single primary melanoma diagnosis and 245 (0.6%) had a secondary primary melanoma diagnosis. The researchers found that second melanomas were more likely than first melanomas to be in situ; for invasive tumors, second melanomas were more likely to be thin (defined as 1 mm or less) than first melanomas. In addition, having many self-reported moles at age 21 years was more strongly associated with second melanomas compared with first melanomas (hazard ratio, 6.36 vs 3.46, respectively; P = .01), as was having a high genetic predisposition (HR, 3.28 vs. 2.06; P = .03). Second melanomas were also more strongly associated with a history of multiple skin cancer excisions (HR, 2.63 vs 1.86; P = .05).

Dr. Kent noted that while sunscreen use protects against melanoma, a National Ambulatory Medical Care Survey in 2014 found that internists and pediatricians mentioned sunscreen at fewer than 0.1% of visits — even those with patients who have a diagnosis of skin disease. “Physicians need to do better,” he said at the meeting, which was hosted by the Scripps Cancer Center. “We as dermatologists have work to do to help educate them.”

Dr. Kent reported having no relevant disclosures.

SAN DIEGO — According to data from the American Cancer Society (ACS), cutaneous melanoma was the fifth most common cancer in 2023, with an estimated 97,610 new cases and 7,990 deaths, following cancer of the colorectal area, lung and bronchus, prostate, and breast.

“The incidence of melanoma seems to have continued to go up since the early 1990s,” David E. Kent, MD, a dermatologist at Skin Care Physicians of Georgia, Macon, said at the annual Cutaneous Malignancy Update. “The death rates have been flat and may have slightly decreased.”

In 2024, the ACS estimates that about 100,640 new melanomas will be diagnosed in the United States (59,170 in men and 41,470 in women), and about 8,290 people are expected to die of melanoma (5,430 men and 2,860 women). Meanwhile, the lifetime risk of melanoma is about 3% (1 in 33) for Whites, 0.1% (1 in 1,000) for Blacks, and 0.5% (1 in 200) for Hispanics. In 2019, there were an estimated 1.4 million people in the United States living with cutaneous melanoma, and the overall 5-year survival is 93.7%.

Epidemiologic studies show an increase in melanoma incidence, primarily among White populations. “This is believed to be due to sun exposure, changes in recreational behaviors, and tanning bed exposures,” said Dr. Kent, who holds a faculty position in the department of dermatology at the Medical College of Georgia, Augusta. Increased surveillance and diagnosis also play a role. In the medical literature, annual increases in melanoma incidence vary from 3%-7% per year, “which means that the rate is doubling every 10-20 years,” he said, noting that annual melanoma costs are approximately $3.3 billion.

While incidence rates are lower in non-White, non-Hispanic populations, poor outcomes are disproportionately higher in persons of color, according to a 2023 paper. Black individuals present at diagnosis with more advanced stage disease and are 1.5 times more likely to die from melanoma, he said, while Hispanics are 2.4 times more likely to present with stage III disease and 3.6 times more likely to have distant metastases. Persons of color also have higher rates of mucosal, acral lentiginous, and subungual melanoma.

Risk Factors

Known genetic risk factors for melanoma include having skin types I and II, particularly those with light hair, light eyes, and freckling, and those with a family history have a twofold increased risk. Also, up to 40% of genetic cases are from inherited mutations in CDKN2A, CDK4, BAP1 and MCR1. Other genetic-related risk factors include the number and size of nevi, having atypical nevus syndrome, DNA repair defects, large congenital nevi, and a personal history of melanoma.

Dr. Kent said that genetic testing for melanoma is warranted for individuals who meet criteria for the “rule of 3s.” He defined this as three primary melanomas in an individual, or three cases of melanoma in first- or second-degree relatives, or two cases of melanoma and one pancreatic cancer or astrocytoma in first or second-degree relatives, or one case of melanoma and two of pancreatic cancer/astrocytoma in first- or second-degree relatives.

The main environmental risk factor for melanoma is exposure to ultraviolet radiation. Individuals can track their UV exposure with a variety of wearable devices and apps, including SunSense One Digital UV Tracker, the SunSense App, the UV Index Widget, the SunSmart Global UV App, the SunKnown UV light photometer, and the EPA’s UV Index Mobile App. Other environment-related risk factors include having a high socioeconomic status (SES), being immunosuppressed, as well as exposure to heavy metals, insecticides, or hormones; and ones distance from the equator.

In a study published in 2023, researchers investigated the risk factors associated with first and second primary melanomas in 38,845 patients who were followed in Australia between 2011 and 2018. During a median follow-up of 7.4 years, 1212 patients (3.1%) had a single primary melanoma diagnosis and 245 (0.6%) had a secondary primary melanoma diagnosis. The researchers found that second melanomas were more likely than first melanomas to be in situ; for invasive tumors, second melanomas were more likely to be thin (defined as 1 mm or less) than first melanomas. In addition, having many self-reported moles at age 21 years was more strongly associated with second melanomas compared with first melanomas (hazard ratio, 6.36 vs 3.46, respectively; P = .01), as was having a high genetic predisposition (HR, 3.28 vs. 2.06; P = .03). Second melanomas were also more strongly associated with a history of multiple skin cancer excisions (HR, 2.63 vs 1.86; P = .05).

Dr. Kent noted that while sunscreen use protects against melanoma, a National Ambulatory Medical Care Survey in 2014 found that internists and pediatricians mentioned sunscreen at fewer than 0.1% of visits — even those with patients who have a diagnosis of skin disease. “Physicians need to do better,” he said at the meeting, which was hosted by the Scripps Cancer Center. “We as dermatologists have work to do to help educate them.”

Dr. Kent reported having no relevant disclosures.

SAN DIEGO — According to data from the American Cancer Society (ACS), cutaneous melanoma was the fifth most common cancer in 2023, with an estimated 97,610 new cases and 7,990 deaths, following cancer of the colorectal area, lung and bronchus, prostate, and breast.

“The incidence of melanoma seems to have continued to go up since the early 1990s,” David E. Kent, MD, a dermatologist at Skin Care Physicians of Georgia, Macon, said at the annual Cutaneous Malignancy Update. “The death rates have been flat and may have slightly decreased.”

In 2024, the ACS estimates that about 100,640 new melanomas will be diagnosed in the United States (59,170 in men and 41,470 in women), and about 8,290 people are expected to die of melanoma (5,430 men and 2,860 women). Meanwhile, the lifetime risk of melanoma is about 3% (1 in 33) for Whites, 0.1% (1 in 1,000) for Blacks, and 0.5% (1 in 200) for Hispanics. In 2019, there were an estimated 1.4 million people in the United States living with cutaneous melanoma, and the overall 5-year survival is 93.7%.

Epidemiologic studies show an increase in melanoma incidence, primarily among White populations. “This is believed to be due to sun exposure, changes in recreational behaviors, and tanning bed exposures,” said Dr. Kent, who holds a faculty position in the department of dermatology at the Medical College of Georgia, Augusta. Increased surveillance and diagnosis also play a role. In the medical literature, annual increases in melanoma incidence vary from 3%-7% per year, “which means that the rate is doubling every 10-20 years,” he said, noting that annual melanoma costs are approximately $3.3 billion.

While incidence rates are lower in non-White, non-Hispanic populations, poor outcomes are disproportionately higher in persons of color, according to a 2023 paper. Black individuals present at diagnosis with more advanced stage disease and are 1.5 times more likely to die from melanoma, he said, while Hispanics are 2.4 times more likely to present with stage III disease and 3.6 times more likely to have distant metastases. Persons of color also have higher rates of mucosal, acral lentiginous, and subungual melanoma.

Risk Factors

Known genetic risk factors for melanoma include having skin types I and II, particularly those with light hair, light eyes, and freckling, and those with a family history have a twofold increased risk. Also, up to 40% of genetic cases are from inherited mutations in CDKN2A, CDK4, BAP1 and MCR1. Other genetic-related risk factors include the number and size of nevi, having atypical nevus syndrome, DNA repair defects, large congenital nevi, and a personal history of melanoma.

Dr. Kent said that genetic testing for melanoma is warranted for individuals who meet criteria for the “rule of 3s.” He defined this as three primary melanomas in an individual, or three cases of melanoma in first- or second-degree relatives, or two cases of melanoma and one pancreatic cancer or astrocytoma in first or second-degree relatives, or one case of melanoma and two of pancreatic cancer/astrocytoma in first- or second-degree relatives.

The main environmental risk factor for melanoma is exposure to ultraviolet radiation. Individuals can track their UV exposure with a variety of wearable devices and apps, including SunSense One Digital UV Tracker, the SunSense App, the UV Index Widget, the SunSmart Global UV App, the SunKnown UV light photometer, and the EPA’s UV Index Mobile App. Other environment-related risk factors include having a high socioeconomic status (SES), being immunosuppressed, as well as exposure to heavy metals, insecticides, or hormones; and ones distance from the equator.

In a study published in 2023, researchers investigated the risk factors associated with first and second primary melanomas in 38,845 patients who were followed in Australia between 2011 and 2018. During a median follow-up of 7.4 years, 1212 patients (3.1%) had a single primary melanoma diagnosis and 245 (0.6%) had a secondary primary melanoma diagnosis. The researchers found that second melanomas were more likely than first melanomas to be in situ; for invasive tumors, second melanomas were more likely to be thin (defined as 1 mm or less) than first melanomas. In addition, having many self-reported moles at age 21 years was more strongly associated with second melanomas compared with first melanomas (hazard ratio, 6.36 vs 3.46, respectively; P = .01), as was having a high genetic predisposition (HR, 3.28 vs. 2.06; P = .03). Second melanomas were also more strongly associated with a history of multiple skin cancer excisions (HR, 2.63 vs 1.86; P = .05).

Dr. Kent noted that while sunscreen use protects against melanoma, a National Ambulatory Medical Care Survey in 2014 found that internists and pediatricians mentioned sunscreen at fewer than 0.1% of visits — even those with patients who have a diagnosis of skin disease. “Physicians need to do better,” he said at the meeting, which was hosted by the Scripps Cancer Center. “We as dermatologists have work to do to help educate them.”

Dr. Kent reported having no relevant disclosures.

TOPLINE:

Melanoma in situ drives most cases of melanoma overdiagnoses, according to an analysis by Surveillance, Epidemiology, and End Results (SEER).

METHODOLOGY:

• The increase in melanoma diagnoses in the United States, while mortality has remained flat, has raised concerns about overdiagnosis of melanoma, cases that may not result in harm if left untreated. How much of the overdiagnoses can be attributed to melanoma in situ vs invasive melanoma is unknown.
• To address this question, researchers collected data from the SEER 9 registries database.
• They used DevCan software to calculate the cumulative lifetime risk of White American men and women being diagnosed with melanoma between 1975 and 2018, adjusting for changes in longevity and risk factors over the study period.
• The primary outcome was excess lifetime risk for melanoma diagnosis between 1976 and 2018, adjusted for year 2018 competing mortality and changes in risk factors.

TAKEAWAY:

• Researchers found that between 1975 and 2018, the adjusted lifetime risk of being diagnosed with melanoma in situ increased from 0.17% to 2.7% in White men and 0.08% to 2% in White women.
• An estimated 49.7% and 64.6% of melanomas diagnosed in White men and White women, respectively, were overdiagnosed in 2018.
• Among individuals diagnosed with melanoma in situ, 89.4% of White men and 85.4% of White women were likely overdiagnosed in 2018.

IN PRACTICE:

“A large proportion of overdiagnosed melanomas are in situ cancers, pointing to a potential area to focus for an intervention de-escalation of the intensity of treatment and survivorship care,” the authors wrote.

SOURCE:

Adewole S. Adamson, MD, of the Division of Dermatology at The University of Texas at Austin Dell Medical School, led the research. The study was published in BMJ Evidence-Based Medicine on January 19, 2024.

LIMITATIONS:

The analysis only involved White individuals. Other limitations include a high risk for selection bias and that the researchers assumed no melanoma diagnosis in 1975, which may not be the case.

DISCLOSURES:

Dr. Adamson disclosed that he is supported by the Robert Wood Johnson Foundation through The Harold Amos Medical Faculty Development Program. Coauthor Katy J.L. Bell, MBchB, PhD, of the University of Sydney, is supported by an Australian Government National Health and Medical Research Council Investigator Grant.

A version of this article first appeared on Medscape.com.

TOPLINE:

Melanoma in situ drives most cases of melanoma overdiagnoses, according to an analysis by Surveillance, Epidemiology, and End Results (SEER).

METHODOLOGY:

• The increase in melanoma diagnoses in the United States, while mortality has remained flat, has raised concerns about overdiagnosis of melanoma, cases that may not result in harm if left untreated. How much of the overdiagnoses can be attributed to melanoma in situ vs invasive melanoma is unknown.
• To address this question, researchers collected data from the SEER 9 registries database.
• They used DevCan software to calculate the cumulative lifetime risk of White American men and women being diagnosed with melanoma between 1975 and 2018, adjusting for changes in longevity and risk factors over the study period.
• The primary outcome was excess lifetime risk for melanoma diagnosis between 1976 and 2018, adjusted for year 2018 competing mortality and changes in risk factors.

TAKEAWAY:

• Researchers found that between 1975 and 2018, the adjusted lifetime risk of being diagnosed with melanoma in situ increased from 0.17% to 2.7% in White men and 0.08% to 2% in White women.
• An estimated 49.7% and 64.6% of melanomas diagnosed in White men and White women, respectively, were overdiagnosed in 2018.
• Among individuals diagnosed with melanoma in situ, 89.4% of White men and 85.4% of White women were likely overdiagnosed in 2018.

IN PRACTICE:

“A large proportion of overdiagnosed melanomas are in situ cancers, pointing to a potential area to focus for an intervention de-escalation of the intensity of treatment and survivorship care,” the authors wrote.

SOURCE:

Adewole S. Adamson, MD, of the Division of Dermatology at The University of Texas at Austin Dell Medical School, led the research. The study was published in BMJ Evidence-Based Medicine on January 19, 2024.

LIMITATIONS:

The analysis only involved White individuals. Other limitations include a high risk for selection bias and that the researchers assumed no melanoma diagnosis in 1975, which may not be the case.

DISCLOSURES:

Dr. Adamson disclosed that he is supported by the Robert Wood Johnson Foundation through The Harold Amos Medical Faculty Development Program. Coauthor Katy J.L. Bell, MBchB, PhD, of the University of Sydney, is supported by an Australian Government National Health and Medical Research Council Investigator Grant.

A version of this article first appeared on Medscape.com.

TOPLINE:

Melanoma in situ drives most cases of melanoma overdiagnoses, according to an analysis by Surveillance, Epidemiology, and End Results (SEER).

METHODOLOGY:

• The increase in melanoma diagnoses in the United States, while mortality has remained flat, has raised concerns about overdiagnosis of melanoma, cases that may not result in harm if left untreated. How much of the overdiagnoses can be attributed to melanoma in situ vs invasive melanoma is unknown.
• To address this question, researchers collected data from the SEER 9 registries database.
• They used DevCan software to calculate the cumulative lifetime risk of White American men and women being diagnosed with melanoma between 1975 and 2018, adjusting for changes in longevity and risk factors over the study period.
• The primary outcome was excess lifetime risk for melanoma diagnosis between 1976 and 2018, adjusted for year 2018 competing mortality and changes in risk factors.

TAKEAWAY:

• Researchers found that between 1975 and 2018, the adjusted lifetime risk of being diagnosed with melanoma in situ increased from 0.17% to 2.7% in White men and 0.08% to 2% in White women.
• An estimated 49.7% and 64.6% of melanomas diagnosed in White men and White women, respectively, were overdiagnosed in 2018.
• Among individuals diagnosed with melanoma in situ, 89.4% of White men and 85.4% of White women were likely overdiagnosed in 2018.

IN PRACTICE:

“A large proportion of overdiagnosed melanomas are in situ cancers, pointing to a potential area to focus for an intervention de-escalation of the intensity of treatment and survivorship care,” the authors wrote.

SOURCE:

Adewole S. Adamson, MD, of the Division of Dermatology at The University of Texas at Austin Dell Medical School, led the research. The study was published in BMJ Evidence-Based Medicine on January 19, 2024.

LIMITATIONS:

The analysis only involved White individuals. Other limitations include a high risk for selection bias and that the researchers assumed no melanoma diagnosis in 1975, which may not be the case.

DISCLOSURES:

Dr. Adamson disclosed that he is supported by the Robert Wood Johnson Foundation through The Harold Amos Medical Faculty Development Program. Coauthor Katy J.L. Bell, MBchB, PhD, of the University of Sydney, is supported by an Australian Government National Health and Medical Research Council Investigator Grant.

A version of this article first appeared on Medscape.com.

In April 2023, the US Preventive Services Task Force (USPSTF) issued a controversial recommendation that the current evidence is insufficient to assess the benefits vs harms of visual skin examination by clinicians for skin cancer screening in adolescents and adults who do not have signs or symptoms of skin cancer.^1,2 This recommendation by the USPSTF has not changed in a quarter century,³ but a recent study described an interesting paradox that should trigger wide evaluation and debate.

Patel et al⁴ analyzed data from the National Cancer Institute’s Surveillance, Epidemiology, and End Results Program from January 2000 to December 2018 to identify adults with a diagnosis of first primary melanoma in situ (MIS). Overall mortality was then determined through the National Vital Statistics System, which provides cause-of-death information for all deaths in the United States. The authors found 137,872 patients who had 1—and only 1—MIS discovered over the observation period. These patients predominantly were White (96.7%), and the mean (SD) age at diagnosis was 61.9 (16.5) years. During 910,308 total person-years of follow-up (mean [SD], 6.6 [5.1] years), 893 (0.6%) patients died of melanoma and 17,327 (12.6%) died of any cause. The 15-year melanoma-specific standardized mortality rate (SMR) was 1.89 (95% CI, 1.77-2.02), yet the 15-year overall survival relative to matched population controls was 112.4% (95% CI, 112.0%-112.8%), thus all-cause SMR was significantly lower at 0.68 (95% CI, 0.67-0.7). Although MIS was associated with a small increase in cohort melanoma mortality, overall mortality was actually lower than in the general population.⁴

Patel et al⁴ did a further broader search that included an additional 18,379 patients who also experienced a second primary melanoma, of which 6751 (36.7%) were invasive and 11,628 (63.3%) were in situ, with a melanoma-specific survival of 98.2% (95% CI, 97.6%-98.5%). Yet relative all-cause survival was significantly higher at 126.7% (95% CI, 125.5%-128.0%). Even among patients in whom a second primary melanoma was invasive, melanoma-specific survival was reduced to 91.1% (95% CI, 90.0%-92.1%), but relative all-cause survival was 116.7% (95% CI, 115%-118.4%). These data in the overall cohort of 155,251 patients showed a discordance between melanoma mortality, which was 4.27-times higher than in the general population (SMR, 4.27; 95% CI, 4.07-4.48), and a lower risk for death from all causes that was approximately 27% lower than in the general population (SMR, 0.73; 95% CI, 0.72-0.74). The authors showed that their findings were not associated with socioeconomic status.⁴

The analysis by Patel et al⁴ is now the second study in the literature to show this discordant melanoma survival pattern. In an earlier Australian study of 2452 melanoma patients, Watts et al⁵ reported that melanoma detection during routine skin checks was associated with a 25% lower all-cause mortality (hazard ratio, 0.75; 95% CI, 0.63-0.90) but not melanoma-specific mortality after multivariable adjustment for a variety of factors including socioeconomic status.These analyses by 2 different groups of investigators have broad implications. Both groups suggested that the improved life span in melanoma patients may be due to health-seeking behavior, which has been defined as “any action undertaken by individuals who perceive themselves to have a health problem or to be ill for the purpose of finding an appropriate remedy.”⁶

Once treated for melanoma, it is clear that patients are likely to return at regular intervals for thorough full-body skin examinations, but this activity alone could not be responsible for improved all-cause mortality in the face of increased melanoma-specific mortality. It seems the authors are implying a broader concept of good health behavior, originally defined by MacKian⁷ as encompassing “activities undertaken to maintain good health, to prevent ill health, as well as dealing with any departure from a good state of health,” such as overt behavioral patterns, actions, and habits with the goal of maintenance, restoration, and improvement of one’s health. A variety of behaviors fall within such a definition including smoking cessation, decreased alcohol use, good diet, more physical activity, safe sexual behavior, scheduling physician visits, medication adherence, vaccination, and yes—screening examinations for health problems.⁸

The concept that individuals who are diagnosed with melanoma fall into a pattern of good health behavior is an interesting hypothesis that must remain speculative until the multiple aspects of good health behavior are rigorously studied. This concept coexists with the hypothesis of melanoma “overdiagnosis”—the idea that many melanomas are detected that will never lead to death.⁹ Both concepts deserve further analysis. Unquestionably, a randomized controlled trial could never recruit patients willing to undergo long-term untreated observation of their melanomas to test the hypothesis that their melanoma diagnosis would eventually lead to death. Furthermore, Patel et al⁴ do suggest that even MIS carries a small but measurable increased risk for death from the disease, which is not particularly supportive of the overdiagnosis hypothesis; however, analysis of the concept that improved individual health behavior is at least in part responsible for the first discovery of melanomas is certainly approachable. Here is the key question: Did the melanoma diagnosis trigger a sudden change in multiple aspects of health behavior that led to significant all-cause mortality benefits? The average age of the population studied by Patel et al⁴ was approximately 62 years. One wonders whether the consequences of a lifetime of established health behavior patterns can be rapidly ­modified—certainly possible but again remains to be proven by further studies.

Conversely, the alternative hypothesis is that discovery of MIS was the result of active pursuit of self-examination and screening procedures as part of individually ingrained good health behavior over a lifetime. Goodwin et al¹⁰ carried out a study in a sample of the Medicare population aged 69 to 90 years looking at men who had prostate cancer screening via prostate-specific antigen measurement and women who had undergone mammography in older age, compared to the contrast population who had not had these screening procedures. They tracked date of death in Medicare enrollment files. They identified 543,970 women and 362,753 men who were aged 69 to 90 years as of January 1, 2003. Patients were stratified by life expectancy based on age and comorbidity. Within each stratum, the patients with cancer screening had higher actual median survival than those who were not screened, with differences ranging from 1.7 to 2.1 years for women and 0.9 to 1.1 years for men.¹⁰ These results were not the result of lower prostate or breast cancer mortality. Rather, one surmises that other health factors yielded lower mortality in the screened cohorts.

A full-body skin examination is a time-consuming process. Patients who come to their physician for a routine annual physical don’t expect a skin examination and very few physicians have the time for a long detailed full-body skin examination. When the patient presents to a dermatologist for an examination, it often is because they have real concerns; for example, they may have had a family member who died of skin cancer, or the patient themself may have noticed a worrisome lesion. Patients, not physicians, are the drivers of skin cancer screening, a fact that often is dismissed by those who are not necessarily supportive of the practice.

In light of the findings of Patel et al,⁴ it is essential that the USPSTF reviews be reanalyzed to compare skin cancer–specific mortality, all-cause mortality, and lifespan in individuals who pursue skin cancer screening; the reanalysis also should not be exclusively limited to survival. With the advent of the immune checkpoint inhibitors, patients with metastatic melanoma are living much longer.¹¹ The burden of living with metastatic cancer must be characterized and measured to have a complete picture and a valid analysis.

After the release of the USPSTF recommendation, there have been calls for large-scale studies to prove the benefits of skin cancer screening.¹² Such studies may be valuable; however, if the hypothesis that overall healthy behavior as the major outcome determinant is substantiated, it may prove quite challenging to perform tests of association with specific interventions. It has been shown that skin cancer screening does lead to discovery of more melanomas,¹³ yet in light of the paradox described by Patel et al,⁴ it also is likely that causes of death other than melanoma impact overall mortality. Patients who pursue skin examinations may engage in multiple different health activities that are beneficial in the long term, making it difficult to analyze the specific benefit of skin cancer screening in isolation. It may prove difficult to ask patients to omit selected aspects of healthy behavior to try to prove the point. At this time, there is much more work to be done prior to offering opinions on the importance of skin cancer examination in isolation to improve overall health care. In the meantime, dermatologists owe it to our patients to continue to diligently pursue thorough and detailed skin examinations.

In April 2023, the US Preventive Services Task Force (USPSTF) issued a controversial recommendation that the current evidence is insufficient to assess the benefits vs harms of visual skin examination by clinicians for skin cancer screening in adolescents and adults who do not have signs or symptoms of skin cancer.^1,2 This recommendation by the USPSTF has not changed in a quarter century,³ but a recent study described an interesting paradox that should trigger wide evaluation and debate.

Patel et al⁴ analyzed data from the National Cancer Institute’s Surveillance, Epidemiology, and End Results Program from January 2000 to December 2018 to identify adults with a diagnosis of first primary melanoma in situ (MIS). Overall mortality was then determined through the National Vital Statistics System, which provides cause-of-death information for all deaths in the United States. The authors found 137,872 patients who had 1—and only 1—MIS discovered over the observation period. These patients predominantly were White (96.7%), and the mean (SD) age at diagnosis was 61.9 (16.5) years. During 910,308 total person-years of follow-up (mean [SD], 6.6 [5.1] years), 893 (0.6%) patients died of melanoma and 17,327 (12.6%) died of any cause. The 15-year melanoma-specific standardized mortality rate (SMR) was 1.89 (95% CI, 1.77-2.02), yet the 15-year overall survival relative to matched population controls was 112.4% (95% CI, 112.0%-112.8%), thus all-cause SMR was significantly lower at 0.68 (95% CI, 0.67-0.7). Although MIS was associated with a small increase in cohort melanoma mortality, overall mortality was actually lower than in the general population.⁴

Patel et al⁴ did a further broader search that included an additional 18,379 patients who also experienced a second primary melanoma, of which 6751 (36.7%) were invasive and 11,628 (63.3%) were in situ, with a melanoma-specific survival of 98.2% (95% CI, 97.6%-98.5%). Yet relative all-cause survival was significantly higher at 126.7% (95% CI, 125.5%-128.0%). Even among patients in whom a second primary melanoma was invasive, melanoma-specific survival was reduced to 91.1% (95% CI, 90.0%-92.1%), but relative all-cause survival was 116.7% (95% CI, 115%-118.4%). These data in the overall cohort of 155,251 patients showed a discordance between melanoma mortality, which was 4.27-times higher than in the general population (SMR, 4.27; 95% CI, 4.07-4.48), and a lower risk for death from all causes that was approximately 27% lower than in the general population (SMR, 0.73; 95% CI, 0.72-0.74). The authors showed that their findings were not associated with socioeconomic status.⁴

The analysis by Patel et al⁴ is now the second study in the literature to show this discordant melanoma survival pattern. In an earlier Australian study of 2452 melanoma patients, Watts et al⁵ reported that melanoma detection during routine skin checks was associated with a 25% lower all-cause mortality (hazard ratio, 0.75; 95% CI, 0.63-0.90) but not melanoma-specific mortality after multivariable adjustment for a variety of factors including socioeconomic status.These analyses by 2 different groups of investigators have broad implications. Both groups suggested that the improved life span in melanoma patients may be due to health-seeking behavior, which has been defined as “any action undertaken by individuals who perceive themselves to have a health problem or to be ill for the purpose of finding an appropriate remedy.”⁶

Once treated for melanoma, it is clear that patients are likely to return at regular intervals for thorough full-body skin examinations, but this activity alone could not be responsible for improved all-cause mortality in the face of increased melanoma-specific mortality. It seems the authors are implying a broader concept of good health behavior, originally defined by MacKian⁷ as encompassing “activities undertaken to maintain good health, to prevent ill health, as well as dealing with any departure from a good state of health,” such as overt behavioral patterns, actions, and habits with the goal of maintenance, restoration, and improvement of one’s health. A variety of behaviors fall within such a definition including smoking cessation, decreased alcohol use, good diet, more physical activity, safe sexual behavior, scheduling physician visits, medication adherence, vaccination, and yes—screening examinations for health problems.⁸

The concept that individuals who are diagnosed with melanoma fall into a pattern of good health behavior is an interesting hypothesis that must remain speculative until the multiple aspects of good health behavior are rigorously studied. This concept coexists with the hypothesis of melanoma “overdiagnosis”—the idea that many melanomas are detected that will never lead to death.⁹ Both concepts deserve further analysis. Unquestionably, a randomized controlled trial could never recruit patients willing to undergo long-term untreated observation of their melanomas to test the hypothesis that their melanoma diagnosis would eventually lead to death. Furthermore, Patel et al⁴ do suggest that even MIS carries a small but measurable increased risk for death from the disease, which is not particularly supportive of the overdiagnosis hypothesis; however, analysis of the concept that improved individual health behavior is at least in part responsible for the first discovery of melanomas is certainly approachable. Here is the key question: Did the melanoma diagnosis trigger a sudden change in multiple aspects of health behavior that led to significant all-cause mortality benefits? The average age of the population studied by Patel et al⁴ was approximately 62 years. One wonders whether the consequences of a lifetime of established health behavior patterns can be rapidly ­modified—certainly possible but again remains to be proven by further studies.

Conversely, the alternative hypothesis is that discovery of MIS was the result of active pursuit of self-examination and screening procedures as part of individually ingrained good health behavior over a lifetime. Goodwin et al¹⁰ carried out a study in a sample of the Medicare population aged 69 to 90 years looking at men who had prostate cancer screening via prostate-specific antigen measurement and women who had undergone mammography in older age, compared to the contrast population who had not had these screening procedures. They tracked date of death in Medicare enrollment files. They identified 543,970 women and 362,753 men who were aged 69 to 90 years as of January 1, 2003. Patients were stratified by life expectancy based on age and comorbidity. Within each stratum, the patients with cancer screening had higher actual median survival than those who were not screened, with differences ranging from 1.7 to 2.1 years for women and 0.9 to 1.1 years for men.¹⁰ These results were not the result of lower prostate or breast cancer mortality. Rather, one surmises that other health factors yielded lower mortality in the screened cohorts.

A full-body skin examination is a time-consuming process. Patients who come to their physician for a routine annual physical don’t expect a skin examination and very few physicians have the time for a long detailed full-body skin examination. When the patient presents to a dermatologist for an examination, it often is because they have real concerns; for example, they may have had a family member who died of skin cancer, or the patient themself may have noticed a worrisome lesion. Patients, not physicians, are the drivers of skin cancer screening, a fact that often is dismissed by those who are not necessarily supportive of the practice.

In light of the findings of Patel et al,⁴ it is essential that the USPSTF reviews be reanalyzed to compare skin cancer–specific mortality, all-cause mortality, and lifespan in individuals who pursue skin cancer screening; the reanalysis also should not be exclusively limited to survival. With the advent of the immune checkpoint inhibitors, patients with metastatic melanoma are living much longer.¹¹ The burden of living with metastatic cancer must be characterized and measured to have a complete picture and a valid analysis.

After the release of the USPSTF recommendation, there have been calls for large-scale studies to prove the benefits of skin cancer screening.¹² Such studies may be valuable; however, if the hypothesis that overall healthy behavior as the major outcome determinant is substantiated, it may prove quite challenging to perform tests of association with specific interventions. It has been shown that skin cancer screening does lead to discovery of more melanomas,¹³ yet in light of the paradox described by Patel et al,⁴ it also is likely that causes of death other than melanoma impact overall mortality. Patients who pursue skin examinations may engage in multiple different health activities that are beneficial in the long term, making it difficult to analyze the specific benefit of skin cancer screening in isolation. It may prove difficult to ask patients to omit selected aspects of healthy behavior to try to prove the point. At this time, there is much more work to be done prior to offering opinions on the importance of skin cancer examination in isolation to improve overall health care. In the meantime, dermatologists owe it to our patients to continue to diligently pursue thorough and detailed skin examinations.

In April 2023, the US Preventive Services Task Force (USPSTF) issued a controversial recommendation that the current evidence is insufficient to assess the benefits vs harms of visual skin examination by clinicians for skin cancer screening in adolescents and adults who do not have signs or symptoms of skin cancer.^1,2 This recommendation by the USPSTF has not changed in a quarter century,³ but a recent study described an interesting paradox that should trigger wide evaluation and debate.

Patel et al⁴ analyzed data from the National Cancer Institute’s Surveillance, Epidemiology, and End Results Program from January 2000 to December 2018 to identify adults with a diagnosis of first primary melanoma in situ (MIS). Overall mortality was then determined through the National Vital Statistics System, which provides cause-of-death information for all deaths in the United States. The authors found 137,872 patients who had 1—and only 1—MIS discovered over the observation period. These patients predominantly were White (96.7%), and the mean (SD) age at diagnosis was 61.9 (16.5) years. During 910,308 total person-years of follow-up (mean [SD], 6.6 [5.1] years), 893 (0.6%) patients died of melanoma and 17,327 (12.6%) died of any cause. The 15-year melanoma-specific standardized mortality rate (SMR) was 1.89 (95% CI, 1.77-2.02), yet the 15-year overall survival relative to matched population controls was 112.4% (95% CI, 112.0%-112.8%), thus all-cause SMR was significantly lower at 0.68 (95% CI, 0.67-0.7). Although MIS was associated with a small increase in cohort melanoma mortality, overall mortality was actually lower than in the general population.⁴

Patel et al⁴ did a further broader search that included an additional 18,379 patients who also experienced a second primary melanoma, of which 6751 (36.7%) were invasive and 11,628 (63.3%) were in situ, with a melanoma-specific survival of 98.2% (95% CI, 97.6%-98.5%). Yet relative all-cause survival was significantly higher at 126.7% (95% CI, 125.5%-128.0%). Even among patients in whom a second primary melanoma was invasive, melanoma-specific survival was reduced to 91.1% (95% CI, 90.0%-92.1%), but relative all-cause survival was 116.7% (95% CI, 115%-118.4%). These data in the overall cohort of 155,251 patients showed a discordance between melanoma mortality, which was 4.27-times higher than in the general population (SMR, 4.27; 95% CI, 4.07-4.48), and a lower risk for death from all causes that was approximately 27% lower than in the general population (SMR, 0.73; 95% CI, 0.72-0.74). The authors showed that their findings were not associated with socioeconomic status.⁴

The analysis by Patel et al⁴ is now the second study in the literature to show this discordant melanoma survival pattern. In an earlier Australian study of 2452 melanoma patients, Watts et al⁵ reported that melanoma detection during routine skin checks was associated with a 25% lower all-cause mortality (hazard ratio, 0.75; 95% CI, 0.63-0.90) but not melanoma-specific mortality after multivariable adjustment for a variety of factors including socioeconomic status.These analyses by 2 different groups of investigators have broad implications. Both groups suggested that the improved life span in melanoma patients may be due to health-seeking behavior, which has been defined as “any action undertaken by individuals who perceive themselves to have a health problem or to be ill for the purpose of finding an appropriate remedy.”⁶

Once treated for melanoma, it is clear that patients are likely to return at regular intervals for thorough full-body skin examinations, but this activity alone could not be responsible for improved all-cause mortality in the face of increased melanoma-specific mortality. It seems the authors are implying a broader concept of good health behavior, originally defined by MacKian⁷ as encompassing “activities undertaken to maintain good health, to prevent ill health, as well as dealing with any departure from a good state of health,” such as overt behavioral patterns, actions, and habits with the goal of maintenance, restoration, and improvement of one’s health. A variety of behaviors fall within such a definition including smoking cessation, decreased alcohol use, good diet, more physical activity, safe sexual behavior, scheduling physician visits, medication adherence, vaccination, and yes—screening examinations for health problems.⁸

The concept that individuals who are diagnosed with melanoma fall into a pattern of good health behavior is an interesting hypothesis that must remain speculative until the multiple aspects of good health behavior are rigorously studied. This concept coexists with the hypothesis of melanoma “overdiagnosis”—the idea that many melanomas are detected that will never lead to death.⁹ Both concepts deserve further analysis. Unquestionably, a randomized controlled trial could never recruit patients willing to undergo long-term untreated observation of their melanomas to test the hypothesis that their melanoma diagnosis would eventually lead to death. Furthermore, Patel et al⁴ do suggest that even MIS carries a small but measurable increased risk for death from the disease, which is not particularly supportive of the overdiagnosis hypothesis; however, analysis of the concept that improved individual health behavior is at least in part responsible for the first discovery of melanomas is certainly approachable. Here is the key question: Did the melanoma diagnosis trigger a sudden change in multiple aspects of health behavior that led to significant all-cause mortality benefits? The average age of the population studied by Patel et al⁴ was approximately 62 years. One wonders whether the consequences of a lifetime of established health behavior patterns can be rapidly ­modified—certainly possible but again remains to be proven by further studies.

Conversely, the alternative hypothesis is that discovery of MIS was the result of active pursuit of self-examination and screening procedures as part of individually ingrained good health behavior over a lifetime. Goodwin et al¹⁰ carried out a study in a sample of the Medicare population aged 69 to 90 years looking at men who had prostate cancer screening via prostate-specific antigen measurement and women who had undergone mammography in older age, compared to the contrast population who had not had these screening procedures. They tracked date of death in Medicare enrollment files. They identified 543,970 women and 362,753 men who were aged 69 to 90 years as of January 1, 2003. Patients were stratified by life expectancy based on age and comorbidity. Within each stratum, the patients with cancer screening had higher actual median survival than those who were not screened, with differences ranging from 1.7 to 2.1 years for women and 0.9 to 1.1 years for men.¹⁰ These results were not the result of lower prostate or breast cancer mortality. Rather, one surmises that other health factors yielded lower mortality in the screened cohorts.

A full-body skin examination is a time-consuming process. Patients who come to their physician for a routine annual physical don’t expect a skin examination and very few physicians have the time for a long detailed full-body skin examination. When the patient presents to a dermatologist for an examination, it often is because they have real concerns; for example, they may have had a family member who died of skin cancer, or the patient themself may have noticed a worrisome lesion. Patients, not physicians, are the drivers of skin cancer screening, a fact that often is dismissed by those who are not necessarily supportive of the practice.

In light of the findings of Patel et al,⁴ it is essential that the USPSTF reviews be reanalyzed to compare skin cancer–specific mortality, all-cause mortality, and lifespan in individuals who pursue skin cancer screening; the reanalysis also should not be exclusively limited to survival. With the advent of the immune checkpoint inhibitors, patients with metastatic melanoma are living much longer.¹¹ The burden of living with metastatic cancer must be characterized and measured to have a complete picture and a valid analysis.

After the release of the USPSTF recommendation, there have been calls for large-scale studies to prove the benefits of skin cancer screening.¹² Such studies may be valuable; however, if the hypothesis that overall healthy behavior as the major outcome determinant is substantiated, it may prove quite challenging to perform tests of association with specific interventions. It has been shown that skin cancer screening does lead to discovery of more melanomas,¹³ yet in light of the paradox described by Patel et al,⁴ it also is likely that causes of death other than melanoma impact overall mortality. Patients who pursue skin examinations may engage in multiple different health activities that are beneficial in the long term, making it difficult to analyze the specific benefit of skin cancer screening in isolation. It may prove difficult to ask patients to omit selected aspects of healthy behavior to try to prove the point. At this time, there is much more work to be done prior to offering opinions on the importance of skin cancer examination in isolation to improve overall health care. In the meantime, dermatologists owe it to our patients to continue to diligently pursue thorough and detailed skin examinations.

Officials at Dana-Farber Cancer Institute are moving to retract at least six published research papers and correct 31 others amid allegations of data manipulation.

News of the investigation follows a blog post by British molecular biologist Sholto David, MD, who flagged almost 60 papers published between 1997 and 2017 that contained image manipulation and other errors. Some of the papers were published by Dana-Farber’s chief executive officer, Laurie Glimcher, MD, and chief operating officer, William Hahn, MD, on topics including multiple myeloma and immune cells.

Mr. David, who blogs about research integrity, highlighted numerous errors and irregularities, including copying and pasting images across multiple experiments to represent different days within the same experiment, sometimes rotating or stretching images.

In one case, Mr. David equated the manipulation with tactics used by “hapless Chinese papermills” and concluded that “a swathe of research coming out of [Dana-Farber] authored by the most senior researchers and managers appears to be hopelessly corrupt with errors that are obvious from just a cursory reading the papers.” 

“Imagine what mistakes might be found in the raw data if anyone was allowed to look!” he wrote.

Barrett Rollins, MD, PhD, Dana-Farber Cancer Institute’s research integrity officer, declined to comment on whether the errors represent scientific misconduct, according to STAT. Rollins told ScienceInsider that the “presence of image discrepancies in a paper is not evidence of an author’s intent to deceive.” 

Access to new artificial intelligence tools is making it easier for data sleuths, like Mr. David, to unearth data manipulation and errors. 

The current investigation closely follows two other investigations into the published work of Harvard University’s former president, Claudine Gay, and Stanford University’s former president, Marc Tessier-Lavigne, which led both to resign their posts. 

A version of this article appeared on Medscape.com.

Officials at Dana-Farber Cancer Institute are moving to retract at least six published research papers and correct 31 others amid allegations of data manipulation.

News of the investigation follows a blog post by British molecular biologist Sholto David, MD, who flagged almost 60 papers published between 1997 and 2017 that contained image manipulation and other errors. Some of the papers were published by Dana-Farber’s chief executive officer, Laurie Glimcher, MD, and chief operating officer, William Hahn, MD, on topics including multiple myeloma and immune cells.

Mr. David, who blogs about research integrity, highlighted numerous errors and irregularities, including copying and pasting images across multiple experiments to represent different days within the same experiment, sometimes rotating or stretching images.

In one case, Mr. David equated the manipulation with tactics used by “hapless Chinese papermills” and concluded that “a swathe of research coming out of [Dana-Farber] authored by the most senior researchers and managers appears to be hopelessly corrupt with errors that are obvious from just a cursory reading the papers.” 

“Imagine what mistakes might be found in the raw data if anyone was allowed to look!” he wrote.

Barrett Rollins, MD, PhD, Dana-Farber Cancer Institute’s research integrity officer, declined to comment on whether the errors represent scientific misconduct, according to STAT. Rollins told ScienceInsider that the “presence of image discrepancies in a paper is not evidence of an author’s intent to deceive.” 

Access to new artificial intelligence tools is making it easier for data sleuths, like Mr. David, to unearth data manipulation and errors. 

The current investigation closely follows two other investigations into the published work of Harvard University’s former president, Claudine Gay, and Stanford University’s former president, Marc Tessier-Lavigne, which led both to resign their posts. 

A version of this article appeared on Medscape.com.

Officials at Dana-Farber Cancer Institute are moving to retract at least six published research papers and correct 31 others amid allegations of data manipulation.

News of the investigation follows a blog post by British molecular biologist Sholto David, MD, who flagged almost 60 papers published between 1997 and 2017 that contained image manipulation and other errors. Some of the papers were published by Dana-Farber’s chief executive officer, Laurie Glimcher, MD, and chief operating officer, William Hahn, MD, on topics including multiple myeloma and immune cells.

Mr. David, who blogs about research integrity, highlighted numerous errors and irregularities, including copying and pasting images across multiple experiments to represent different days within the same experiment, sometimes rotating or stretching images.

In one case, Mr. David equated the manipulation with tactics used by “hapless Chinese papermills” and concluded that “a swathe of research coming out of [Dana-Farber] authored by the most senior researchers and managers appears to be hopelessly corrupt with errors that are obvious from just a cursory reading the papers.” 

“Imagine what mistakes might be found in the raw data if anyone was allowed to look!” he wrote.

Barrett Rollins, MD, PhD, Dana-Farber Cancer Institute’s research integrity officer, declined to comment on whether the errors represent scientific misconduct, according to STAT. Rollins told ScienceInsider that the “presence of image discrepancies in a paper is not evidence of an author’s intent to deceive.” 

Access to new artificial intelligence tools is making it easier for data sleuths, like Mr. David, to unearth data manipulation and errors. 

The current investigation closely follows two other investigations into the published work of Harvard University’s former president, Claudine Gay, and Stanford University’s former president, Marc Tessier-Lavigne, which led both to resign their posts. 

A version of this article appeared on Medscape.com.

Radiation oncologists from the largest professional societies have come together to lobby for Medicare payment reform.

The American Society for Radiation Oncology (ASTRO) recently announced its partnership with three other groups — the American College of Radiation Oncology, the American College of Radiology, and the American Society of Clinical Oncology — to change how the specialty is paid for services. 

Over the past decade, radiation oncologists have seen a 23% drop in Medicare reimbursement for radiation therapy services, with more cuts to come, according to a press release from ASTRO.

Traditionally, Medicare has reimbursed on the basis of the fraction of radiation delivered. But with moves toward hypofractionated regimens, deescalated therapy, and other changes in the field, reimbursement has continued to dwindle. 

The cuts have led to practice consolidation and closures that threaten patient access especially in rural and underserved areas, a spokesperson for the group told this news organization.

To reverse this trend, ASTRO recently proposed the Radiation Oncology Case Rate program, a legislative initiative to base reimbursements on patient volumes instead of fractions delivered. 

ASTRO is currently drafting a congressional bill to change the current payment structure, which “has become untenable,” the spokesperson said. 

A version of this article appeared on Medscape.com.

Radiation oncologists from the largest professional societies have come together to lobby for Medicare payment reform.

The American Society for Radiation Oncology (ASTRO) recently announced its partnership with three other groups — the American College of Radiation Oncology, the American College of Radiology, and the American Society of Clinical Oncology — to change how the specialty is paid for services. 

Over the past decade, radiation oncologists have seen a 23% drop in Medicare reimbursement for radiation therapy services, with more cuts to come, according to a press release from ASTRO.

Traditionally, Medicare has reimbursed on the basis of the fraction of radiation delivered. But with moves toward hypofractionated regimens, deescalated therapy, and other changes in the field, reimbursement has continued to dwindle. 

The cuts have led to practice consolidation and closures that threaten patient access especially in rural and underserved areas, a spokesperson for the group told this news organization.

To reverse this trend, ASTRO recently proposed the Radiation Oncology Case Rate program, a legislative initiative to base reimbursements on patient volumes instead of fractions delivered. 

ASTRO is currently drafting a congressional bill to change the current payment structure, which “has become untenable,” the spokesperson said. 

A version of this article appeared on Medscape.com.

Radiation oncologists from the largest professional societies have come together to lobby for Medicare payment reform.

The American Society for Radiation Oncology (ASTRO) recently announced its partnership with three other groups — the American College of Radiation Oncology, the American College of Radiology, and the American Society of Clinical Oncology — to change how the specialty is paid for services. 

Over the past decade, radiation oncologists have seen a 23% drop in Medicare reimbursement for radiation therapy services, with more cuts to come, according to a press release from ASTRO.

Traditionally, Medicare has reimbursed on the basis of the fraction of radiation delivered. But with moves toward hypofractionated regimens, deescalated therapy, and other changes in the field, reimbursement has continued to dwindle. 

The cuts have led to practice consolidation and closures that threaten patient access especially in rural and underserved areas, a spokesperson for the group told this news organization.

To reverse this trend, ASTRO recently proposed the Radiation Oncology Case Rate program, a legislative initiative to base reimbursements on patient volumes instead of fractions delivered. 

ASTRO is currently drafting a congressional bill to change the current payment structure, which “has become untenable,” the spokesperson said. 

A version of this article appeared on Medscape.com.

The Food and Drug Administration has cleared the DermaSensor device for point-of-care, noninvasive testing for all types of skin cancer.

The handheld wireless tool, which was developed by Miami-based DermaSensor Inc., operates on battery power, uses spectroscopy and algorithms to evaluate skin lesions for potential cancer in a matter of seconds, and is intended for use by primary care physicians. After the device completes the scan of a lesion, a result of “investigate further” (positive result) suggests further evaluation through a referral to a dermatologist, while “monitor” (negative result) suggests that there is no immediate need for a referral to a dermatologist.

In a pivotal trial of the device that evaluated 224 high risk lesions at 18 primary care study sites in the United States and 4 in Australia, the device had an overall sensitivity of 95.5% for detecting malignancy.

In a more recent validation study funded by DermaSensor, investigators tested 333 lesions at four U.S. dermatology offices and found that the overall device sensitivity was 97.04%, with subgroup sensitivity of 96.67% for melanoma, 97.22% for basal cell carcinoma, and 97.01% for squamous cell carcinoma. Overall specificity of the device was 26.22%.

The study authors, led by Tallahassee, Fla.–based dermatologist Armand B. Cognetta Jr., MD, concluded that DermaSensor’s rapid clinical analysis of lesions “allows for its easy integration into clinical practice infrastructures. Proper use of this device may aid in the reduction of morbidity and mortality associated with skin cancer through expedited and enhanced detection and intervention.”

According to marketing material from the DermaSensor website, the device’s AI algorithm was developed and validated with more than 20,000 scans, composed of more than 4,000 benign and malignant lesions. In a statement about the clearance, the FDA emphasized that the device “should not be used as the sole diagnostic criterion nor to confirm a diagnosis of skin cancer.” The agency is requiring that the manufacturer “conduct additional post-market clinical validation performance testing of the DermaSensor device in patients from demographic groups representative of the U.S. population, including populations who had limited representation of melanomas in the premarket studies, due to their having a relatively low incidence of the disease.”

According to a spokesperson for DermaSensor, pricing for the device is based on a subscription model: $199 per month for five patients or $399 per month for unlimited use. DermaSensor is currently commercially available in Europe and Australia.

Asked to comment, Vishal A. Patel, MD, director of cutaneous oncology at the George Washington Cancer Center, Washington, said that the FDA clearance of DermaSensor highlights the growing appreciation of AI-driven diagnostic support for primary care providers and dermatologists. "Skin cancers are a growing epidemic in the US and the ability to accurately identify potential suspicious lesions without immediately reaching for the scalpel is invaluable," Patel told this news organization. He was not involved with DermSensor studies.

"Furthermore, this tool can help address the shortage of dermatologists and long wait times by helping primary care providers accurately risk-stratify patients and identify those who need to be seen immediately for potential biopsy and expert care," he added. "However, just like with any new technology, we must use caution to not overutilize this tool," which he said, could "lead to overdiagnosis and overtreatment of early or innocuous lesions that are better managed with empiric field treatments." 

Dr. Cognetta was a paid investigator for the study.

Dr. Patel disclosed that he is chief medical officer for Lazarus AI.

The Food and Drug Administration has cleared the DermaSensor device for point-of-care, noninvasive testing for all types of skin cancer.

The handheld wireless tool, which was developed by Miami-based DermaSensor Inc., operates on battery power, uses spectroscopy and algorithms to evaluate skin lesions for potential cancer in a matter of seconds, and is intended for use by primary care physicians. After the device completes the scan of a lesion, a result of “investigate further” (positive result) suggests further evaluation through a referral to a dermatologist, while “monitor” (negative result) suggests that there is no immediate need for a referral to a dermatologist.

In a pivotal trial of the device that evaluated 224 high risk lesions at 18 primary care study sites in the United States and 4 in Australia, the device had an overall sensitivity of 95.5% for detecting malignancy.

In a more recent validation study funded by DermaSensor, investigators tested 333 lesions at four U.S. dermatology offices and found that the overall device sensitivity was 97.04%, with subgroup sensitivity of 96.67% for melanoma, 97.22% for basal cell carcinoma, and 97.01% for squamous cell carcinoma. Overall specificity of the device was 26.22%.

The study authors, led by Tallahassee, Fla.–based dermatologist Armand B. Cognetta Jr., MD, concluded that DermaSensor’s rapid clinical analysis of lesions “allows for its easy integration into clinical practice infrastructures. Proper use of this device may aid in the reduction of morbidity and mortality associated with skin cancer through expedited and enhanced detection and intervention.”

According to marketing material from the DermaSensor website, the device’s AI algorithm was developed and validated with more than 20,000 scans, composed of more than 4,000 benign and malignant lesions. In a statement about the clearance, the FDA emphasized that the device “should not be used as the sole diagnostic criterion nor to confirm a diagnosis of skin cancer.” The agency is requiring that the manufacturer “conduct additional post-market clinical validation performance testing of the DermaSensor device in patients from demographic groups representative of the U.S. population, including populations who had limited representation of melanomas in the premarket studies, due to their having a relatively low incidence of the disease.”

According to a spokesperson for DermaSensor, pricing for the device is based on a subscription model: $199 per month for five patients or $399 per month for unlimited use. DermaSensor is currently commercially available in Europe and Australia.

Asked to comment, Vishal A. Patel, MD, director of cutaneous oncology at the George Washington Cancer Center, Washington, said that the FDA clearance of DermaSensor highlights the growing appreciation of AI-driven diagnostic support for primary care providers and dermatologists. "Skin cancers are a growing epidemic in the US and the ability to accurately identify potential suspicious lesions without immediately reaching for the scalpel is invaluable," Patel told this news organization. He was not involved with DermSensor studies.

"Furthermore, this tool can help address the shortage of dermatologists and long wait times by helping primary care providers accurately risk-stratify patients and identify those who need to be seen immediately for potential biopsy and expert care," he added. "However, just like with any new technology, we must use caution to not overutilize this tool," which he said, could "lead to overdiagnosis and overtreatment of early or innocuous lesions that are better managed with empiric field treatments." 

Dr. Cognetta was a paid investigator for the study.

Dr. Patel disclosed that he is chief medical officer for Lazarus AI.

The Food and Drug Administration has cleared the DermaSensor device for point-of-care, noninvasive testing for all types of skin cancer.

The handheld wireless tool, which was developed by Miami-based DermaSensor Inc., operates on battery power, uses spectroscopy and algorithms to evaluate skin lesions for potential cancer in a matter of seconds, and is intended for use by primary care physicians. After the device completes the scan of a lesion, a result of “investigate further” (positive result) suggests further evaluation through a referral to a dermatologist, while “monitor” (negative result) suggests that there is no immediate need for a referral to a dermatologist.

In a pivotal trial of the device that evaluated 224 high risk lesions at 18 primary care study sites in the United States and 4 in Australia, the device had an overall sensitivity of 95.5% for detecting malignancy.

In a more recent validation study funded by DermaSensor, investigators tested 333 lesions at four U.S. dermatology offices and found that the overall device sensitivity was 97.04%, with subgroup sensitivity of 96.67% for melanoma, 97.22% for basal cell carcinoma, and 97.01% for squamous cell carcinoma. Overall specificity of the device was 26.22%.

The study authors, led by Tallahassee, Fla.–based dermatologist Armand B. Cognetta Jr., MD, concluded that DermaSensor’s rapid clinical analysis of lesions “allows for its easy integration into clinical practice infrastructures. Proper use of this device may aid in the reduction of morbidity and mortality associated with skin cancer through expedited and enhanced detection and intervention.”

According to marketing material from the DermaSensor website, the device’s AI algorithm was developed and validated with more than 20,000 scans, composed of more than 4,000 benign and malignant lesions. In a statement about the clearance, the FDA emphasized that the device “should not be used as the sole diagnostic criterion nor to confirm a diagnosis of skin cancer.” The agency is requiring that the manufacturer “conduct additional post-market clinical validation performance testing of the DermaSensor device in patients from demographic groups representative of the U.S. population, including populations who had limited representation of melanomas in the premarket studies, due to their having a relatively low incidence of the disease.”

According to a spokesperson for DermaSensor, pricing for the device is based on a subscription model: $199 per month for five patients or $399 per month for unlimited use. DermaSensor is currently commercially available in Europe and Australia.

Asked to comment, Vishal A. Patel, MD, director of cutaneous oncology at the George Washington Cancer Center, Washington, said that the FDA clearance of DermaSensor highlights the growing appreciation of AI-driven diagnostic support for primary care providers and dermatologists. "Skin cancers are a growing epidemic in the US and the ability to accurately identify potential suspicious lesions without immediately reaching for the scalpel is invaluable," Patel told this news organization. He was not involved with DermSensor studies.

"Furthermore, this tool can help address the shortage of dermatologists and long wait times by helping primary care providers accurately risk-stratify patients and identify those who need to be seen immediately for potential biopsy and expert care," he added. "However, just like with any new technology, we must use caution to not overutilize this tool," which he said, could "lead to overdiagnosis and overtreatment of early or innocuous lesions that are better managed with empiric field treatments." 

Dr. Cognetta was a paid investigator for the study.

Dr. Patel disclosed that he is chief medical officer for Lazarus AI.

Moderna and Merck have presented promising results from their phase 2b clinical trial that investigated a combination of a messenger RNA (mRNA) vaccine and a cancer drug for the treatment of melanoma.

Is mRNA set to shake up the world of cancer treatment? This is certainly what Moderna seems to think; the pharmaceutical company has published the results of a phase 2b trial combining its mRNA vaccine (mRNA-4157 [V940]) with Merck’s cancer drug KEYTRUDA. While these are not the final results but rather mid-term data from the 3-year follow-up, they are somewhat promising. The randomized KEYNOTE-942/mRNA-4157-P201 clinical trial involves patients with high-risk (stage III/IV) melanoma following complete resection.

Relapse Risk Halved

Treatment with mRNA-4157 (V940) in combination with pembrolizumab led to a clinically meaningful improvement in recurrence-free survival, reducing the risk for recurrence or death by 49%, compared with pembrolizumab alone. The combination of mRNA-4157 (V940) with pembrolizumab also continued to demonstrate a meaningful improvement in distant metastasis-free survival compared with pembrolizumab alone, reducing the risk of developing distant metastasis or death by 62%. “The KEYNOTE-942/mRNA-4157-P201 study was the first demonstration of efficacy for an investigational mRNA cancer treatment in a randomized clinical trial and the first combination therapy to show a significant benefit over pembrolizumab alone in adjuvant melanoma,” said Kyle Holen, MD, Moderna’s senior vice president, after presenting these results. 

Side Effects

The combined treatment also did not demonstrate more significant side effects than pembrolizumab alone. The number of patients reporting treatment-related adverse events of grade 3 or greater was similar between the arms (25% for mRNA-4157 [V940] with pembrolizumab vs 20% for KEYTRUDA alone). The most common adverse events of any grade attributed to mRNA-4157 (V940) were fatigue (60.6%), injection site pain (56.7%), and chills (49%). Based on data from the phase 2b KEYNOTE-942/mRNA-4157-P201 study, the US Food and Drug Administration and European Medicines Agency granted breakthrough therapy designation and recognition under the the Priority Medicines scheme, respectively, for mRNA-4157 (V940) in combination with KEYTRUDA for the adjuvant treatment of patients with high-risk melanoma.

Phase 3 Trial

In July, Moderna and Merck announced the launch of a phase 3 trial, assessing “mRNA-4157 [V940] in combination with pembrolizumab as adjuvant treatment in patients with high-risk resected melanoma [stages IIB-IV].” Stéphane Bancel, Moderna’s director general, believes that an mRNA vaccine for melanoma could be available in 2025.

Other Cancer Vaccines

Moderna is not the only laboratory to set its sights on developing a vaccine for cancer. In May, BioNTech, in partnership with Roche, proposed a phase 1 clinical trial of a vaccine targeting pancreatic cancer in Nature. In June, at the American Society of Clinical Oncology›s conference, Transgene presented its conclusions concerning its viral vector vaccines against ENT and papillomavirus-linked cancers. And in September, Ose Immunotherapeutics made headlines with its vaccine for advanced lung cancer.

This article was translated from Univadis France, which is part of the Medscape Professional Network.

Moderna and Merck have presented promising results from their phase 2b clinical trial that investigated a combination of a messenger RNA (mRNA) vaccine and a cancer drug for the treatment of melanoma.

Is mRNA set to shake up the world of cancer treatment? This is certainly what Moderna seems to think; the pharmaceutical company has published the results of a phase 2b trial combining its mRNA vaccine (mRNA-4157 [V940]) with Merck’s cancer drug KEYTRUDA. While these are not the final results but rather mid-term data from the 3-year follow-up, they are somewhat promising. The randomized KEYNOTE-942/mRNA-4157-P201 clinical trial involves patients with high-risk (stage III/IV) melanoma following complete resection.

Relapse Risk Halved

Treatment with mRNA-4157 (V940) in combination with pembrolizumab led to a clinically meaningful improvement in recurrence-free survival, reducing the risk for recurrence or death by 49%, compared with pembrolizumab alone. The combination of mRNA-4157 (V940) with pembrolizumab also continued to demonstrate a meaningful improvement in distant metastasis-free survival compared with pembrolizumab alone, reducing the risk of developing distant metastasis or death by 62%. “The KEYNOTE-942/mRNA-4157-P201 study was the first demonstration of efficacy for an investigational mRNA cancer treatment in a randomized clinical trial and the first combination therapy to show a significant benefit over pembrolizumab alone in adjuvant melanoma,” said Kyle Holen, MD, Moderna’s senior vice president, after presenting these results. 

Side Effects

The combined treatment also did not demonstrate more significant side effects than pembrolizumab alone. The number of patients reporting treatment-related adverse events of grade 3 or greater was similar between the arms (25% for mRNA-4157 [V940] with pembrolizumab vs 20% for KEYTRUDA alone). The most common adverse events of any grade attributed to mRNA-4157 (V940) were fatigue (60.6%), injection site pain (56.7%), and chills (49%). Based on data from the phase 2b KEYNOTE-942/mRNA-4157-P201 study, the US Food and Drug Administration and European Medicines Agency granted breakthrough therapy designation and recognition under the the Priority Medicines scheme, respectively, for mRNA-4157 (V940) in combination with KEYTRUDA for the adjuvant treatment of patients with high-risk melanoma.

Phase 3 Trial

In July, Moderna and Merck announced the launch of a phase 3 trial, assessing “mRNA-4157 [V940] in combination with pembrolizumab as adjuvant treatment in patients with high-risk resected melanoma [stages IIB-IV].” Stéphane Bancel, Moderna’s director general, believes that an mRNA vaccine for melanoma could be available in 2025.

Other Cancer Vaccines

Moderna is not the only laboratory to set its sights on developing a vaccine for cancer. In May, BioNTech, in partnership with Roche, proposed a phase 1 clinical trial of a vaccine targeting pancreatic cancer in Nature. In June, at the American Society of Clinical Oncology›s conference, Transgene presented its conclusions concerning its viral vector vaccines against ENT and papillomavirus-linked cancers. And in September, Ose Immunotherapeutics made headlines with its vaccine for advanced lung cancer.

This article was translated from Univadis France, which is part of the Medscape Professional Network.

Moderna and Merck have presented promising results from their phase 2b clinical trial that investigated a combination of a messenger RNA (mRNA) vaccine and a cancer drug for the treatment of melanoma.

Is mRNA set to shake up the world of cancer treatment? This is certainly what Moderna seems to think; the pharmaceutical company has published the results of a phase 2b trial combining its mRNA vaccine (mRNA-4157 [V940]) with Merck’s cancer drug KEYTRUDA. While these are not the final results but rather mid-term data from the 3-year follow-up, they are somewhat promising. The randomized KEYNOTE-942/mRNA-4157-P201 clinical trial involves patients with high-risk (stage III/IV) melanoma following complete resection.

Relapse Risk Halved

Treatment with mRNA-4157 (V940) in combination with pembrolizumab led to a clinically meaningful improvement in recurrence-free survival, reducing the risk for recurrence or death by 49%, compared with pembrolizumab alone. The combination of mRNA-4157 (V940) with pembrolizumab also continued to demonstrate a meaningful improvement in distant metastasis-free survival compared with pembrolizumab alone, reducing the risk of developing distant metastasis or death by 62%. “The KEYNOTE-942/mRNA-4157-P201 study was the first demonstration of efficacy for an investigational mRNA cancer treatment in a randomized clinical trial and the first combination therapy to show a significant benefit over pembrolizumab alone in adjuvant melanoma,” said Kyle Holen, MD, Moderna’s senior vice president, after presenting these results. 

Side Effects

The combined treatment also did not demonstrate more significant side effects than pembrolizumab alone. The number of patients reporting treatment-related adverse events of grade 3 or greater was similar between the arms (25% for mRNA-4157 [V940] with pembrolizumab vs 20% for KEYTRUDA alone). The most common adverse events of any grade attributed to mRNA-4157 (V940) were fatigue (60.6%), injection site pain (56.7%), and chills (49%). Based on data from the phase 2b KEYNOTE-942/mRNA-4157-P201 study, the US Food and Drug Administration and European Medicines Agency granted breakthrough therapy designation and recognition under the the Priority Medicines scheme, respectively, for mRNA-4157 (V940) in combination with KEYTRUDA for the adjuvant treatment of patients with high-risk melanoma.

Phase 3 Trial

In July, Moderna and Merck announced the launch of a phase 3 trial, assessing “mRNA-4157 [V940] in combination with pembrolizumab as adjuvant treatment in patients with high-risk resected melanoma [stages IIB-IV].” Stéphane Bancel, Moderna’s director general, believes that an mRNA vaccine for melanoma could be available in 2025.

Other Cancer Vaccines

Moderna is not the only laboratory to set its sights on developing a vaccine for cancer. In May, BioNTech, in partnership with Roche, proposed a phase 1 clinical trial of a vaccine targeting pancreatic cancer in Nature. In June, at the American Society of Clinical Oncology›s conference, Transgene presented its conclusions concerning its viral vector vaccines against ENT and papillomavirus-linked cancers. And in September, Ose Immunotherapeutics made headlines with its vaccine for advanced lung cancer.

This article was translated from Univadis France, which is part of the Medscape Professional Network.

The assessment and diagnosis of melanocytic lesions can present a formidable challenge to even a seasoned pathologist, which is especially true when dealing with the subset of nevi occurring at special sites—where baseline variations inherent to particular locations on the body can preclude the use of features routinely used to diagnose malignancy elsewhere. These so-called special-site nevi previously have been described in the literature along with suggested criteria for differentiating malignant lesions from their benign counterparts.¹ Locations generally considered to be special sites include the acral skin, anogenital region, breast, ear, and flexural regions.^1,2

When evaluating non–special-site melanocytic lesions, general characteristics associated with a malignant diagnosis include confluence or pagetoid spread of melanocytes, nuclear pleomorphism, cytologic atypia, and irregular architecture³; however, these features can be compatible with a benign diagnosis in special-site nevi depending on their extent and the site in question. Although they can be atypical, special-site nevi tend to have the bulk of their architectural distortion and cytologic atypia in the center of the lesion as opposed to the edges.¹ If a given lesion is from a special site but lacks this reassuring feature, special care should be taken to rule out malignancy.

Preferentially expressed antigen in melanoma (PRAME) is an antigen first identified in tumor-reactive T-cell populations in patients with malignant melanoma. It is the product of an oncogene that frequently is overexpressed in melanomas, lung squamous cell carcinomas, sarcomas, and acute leukemias.⁴ It functions as an antagonist of the retinoic acid signaling pathway, which normally serves to induce further cell differentiation, senescence, or apoptosis.⁵ PRAME inhibits retinoid signaling by forming a complex with both the ligand-bound retinoic acid holoreceptor and the polycomb protein EZH2, which blocks retinoid-dependent gene expression by encouraging chromatin condensation at the RARβ promoter site⁵; therefore, expressing PRAME allows lesional cells a substantial growth advantage.

PRAME expression has been extensively characterized in non–special-site nevi and has filled the need for a rather specific marker of melanoma.^6-10 Although PRAME has been studied in acral nevi,¹¹ the expression pattern in nevi of special sites has yet to be elucidated. Herein, we present a dataset characterizing PRAME expression in these challenging lesions.

Methods

We performed a retrospective case review at the University of Virginia (Charlottesville, Virginia) and collected a panel of 36 special-site nevi that previously were diagnosed as benign by a trained dermatopathologist from January 2020 through December 2022. Special-site nevi were identified using a natural language filter for the following terms: acral, palm, sole, ear, auricular, lip, axilla, armpit, breast, groin, labia, vulva, umbilicus, and penis. This study was approved by the University of Virginia institutional review board.

The original hematoxylin and eosin slides used for primary diagnosis were re-examined to verify the prior diagnosis of benign nevus at a special site. We performed a detailed microscopic examination of all benign nevi in our cohort to determine the frequency of various characteristics at each special site. Sections were prepared from the formalin-fixed and paraffin-embedded tissue blocks and stained with a commercial PRAME antibody (#219650 [Abcam] at a 1:50 dilution) and counterstain. A trained dermatopathologist (S.S.R.) examined the stained sections and recorded the percentage of tumor cells with nuclear PRAME staining. We reported our results using previously established criteria for scoring PRAME immunohistochemistry⁷: 0 for no expression, 1+ for 1% to 25% expression, 2+ for 26% to 50% expression, 3+ for 51% to 75% expression, and 4+ for diffuse or 76% to 100% expression. Only strong clonal expression within a population of cells was graded.

Data handling and statistical testing were performed using the R Project for Statistical Computing (https://www.r-project.org/). Significance testing was performed using the Fisher exact test. Plot construction was performed using ggplot2 (https://ggplot2.tidyverse.org/).

Our study cohort included 36 special-site nevi, and the control cohort comprised 25 melanoma in situ (MIS) or invasive melanoma (IM) lesions occurring at special sites. Table 1 provides a breakdown of the study and control cohorts by lesion site. Table 2 details the results of our microscopic examination, describing frequency of various characteristics of special-site nevi stratified by site.

Of the 36 special-site nevi in our cohort, 20 (56%) had no staining (0) for PRAME, 11 (31%) demonstrated 1+ PRAME expression, 3 (8%) demonstrated 2+ PRAME expression, and 2 (6%) demonstrated 3+ PRAME expression. No nevi showed 4+ expression. In the control cohort, 24 of 25 (96%) MIS and IM showed 3+ or 4+ expression, with 21 (84%) demonstrating ­diffuse/4+ expression. One control case (4%) demonstrated 0 PRAME expression. These data are summarized in Table 3 and Figure 1. There is a significant difference in diffuse (4+) PRAME expression between special-site nevi and MIS/IM occurring at special sites (P=1.039×10^-12).

Based on our cohort, a positivity threshold of 3+ for PRAME expression for the diagnosis of melanoma in a special-site lesion would have a sensitivity of 96% and a specificity of 94%, while a positivity threshold of 4+ for PRAME expression would have a sensitivity of 84% and a specificity of 100%. Figures 2 through 4 show photomicrographs of a special-site nevus of the breast, which appropriately does not stain for PRAME; Figures 5 and 6 show an MIS at a special site that appropriately stains for PRAME.

Comment

The distinction between benign and malignant pigmented lesions at special sites presents a fair challenge for pathologists due to the larger degree of leniency for architectural distortion and cytologic atypia in benign lesions at these sites. The presence of architectural distortion or cytologic atypia at the lesion’s edge makes rendering a benign diagnosis especially difficult, and the need for a validated immunohistochemical stain is apparent. In our cohort, strong clonal PRAME expression provided a reliable immunohistochemical marker, allowing for the distinction of malignant lesions from benign nevi at special sites. Diffuse faint PRAME expression was present in several benign nevi within our cohort, and these lesions were considered negative (0) in our analysis.

Given the described test characteristics, we support the implementation of PRAME immunohistochemistry with a positivity threshold of 4+ expression as an ancillary test supporting the diagnosis of IM or MIS in special sites, which would allow clinicians to leverage the high specificity of 4+ PRAME expression to distinguish an IM or MIS from a benign nevus occurring at a special site. We do not recommend the use of 4+ PRAME expression as a screening test for melanoma or MIS among special-site nevi due to its comparatively low sensitivity; however, no one marker is always reliable, and we recommend continued clinicopathologic correlation for all cases. Although PRAME can assist in the delineation of malignant lesions from benign ones, microscopic examination of hematoxylin and eosin–stained section remains the gold standard for diagnosing malignant melanoma and MIS.

Although our case series included nevi and MIS/IM from all special sites, we were limited in the number of acrogenital and ear nevi included due to a relative paucity of biopsied benign nevi from these locations at the University of Virginia. Additionally, although the magnitude of the difference in PRAME expression between the study and control groups is sufficient to demonstrate statistical significance, the overall strength of our argument would be increased with a larger study group. We were limited by the number of cases available at our institution, which did not utilize PRAME during the initial diagnosis of the case; including these cases in the study group would have undermined the integrity of our argument because the differentiation of benign vs malignant initially was made using PRAME immunohistochemistry.

Conclusion

Due to their atypical features, special-site nevi can be challenging to assess. In this study, we showed that PRAME expression can be a reliable marker to distinguish benign from malignant lesions. Our results showed that 100% of benign special-site nevi demonstrated 3+ expression or less, with 56% (20/36) demonstrating no expression at all. The presence of diffuse PRAME expression (4+ PRAME staining) appears to be a specific indicator of a malignant lesion, but results should always be interpreted with respect to the patient’s clinical history and the lesion’s histomorphologic features. Further study of a larger sample size would allow refinement of the sensitivity and specificity of diffuse PRAME expression in the determination of malignancy for special-site lesions.

Acknowledgment—The authors thank the pathologistsat the University of Virginia Biorepository and Tissue Research Facility (Charlottesville, Virginia) for their skill and expertise in performing immunohistochemical staining for this study.

The assessment and diagnosis of melanocytic lesions can present a formidable challenge to even a seasoned pathologist, which is especially true when dealing with the subset of nevi occurring at special sites—where baseline variations inherent to particular locations on the body can preclude the use of features routinely used to diagnose malignancy elsewhere. These so-called special-site nevi previously have been described in the literature along with suggested criteria for differentiating malignant lesions from their benign counterparts.¹ Locations generally considered to be special sites include the acral skin, anogenital region, breast, ear, and flexural regions.^1,2

When evaluating non–special-site melanocytic lesions, general characteristics associated with a malignant diagnosis include confluence or pagetoid spread of melanocytes, nuclear pleomorphism, cytologic atypia, and irregular architecture³; however, these features can be compatible with a benign diagnosis in special-site nevi depending on their extent and the site in question. Although they can be atypical, special-site nevi tend to have the bulk of their architectural distortion and cytologic atypia in the center of the lesion as opposed to the edges.¹ If a given lesion is from a special site but lacks this reassuring feature, special care should be taken to rule out malignancy.

Preferentially expressed antigen in melanoma (PRAME) is an antigen first identified in tumor-reactive T-cell populations in patients with malignant melanoma. It is the product of an oncogene that frequently is overexpressed in melanomas, lung squamous cell carcinomas, sarcomas, and acute leukemias.⁴ It functions as an antagonist of the retinoic acid signaling pathway, which normally serves to induce further cell differentiation, senescence, or apoptosis.⁵ PRAME inhibits retinoid signaling by forming a complex with both the ligand-bound retinoic acid holoreceptor and the polycomb protein EZH2, which blocks retinoid-dependent gene expression by encouraging chromatin condensation at the RARβ promoter site⁵; therefore, expressing PRAME allows lesional cells a substantial growth advantage.

PRAME expression has been extensively characterized in non–special-site nevi and has filled the need for a rather specific marker of melanoma.^6-10 Although PRAME has been studied in acral nevi,¹¹ the expression pattern in nevi of special sites has yet to be elucidated. Herein, we present a dataset characterizing PRAME expression in these challenging lesions.

Methods

We performed a retrospective case review at the University of Virginia (Charlottesville, Virginia) and collected a panel of 36 special-site nevi that previously were diagnosed as benign by a trained dermatopathologist from January 2020 through December 2022. Special-site nevi were identified using a natural language filter for the following terms: acral, palm, sole, ear, auricular, lip, axilla, armpit, breast, groin, labia, vulva, umbilicus, and penis. This study was approved by the University of Virginia institutional review board.

The original hematoxylin and eosin slides used for primary diagnosis were re-examined to verify the prior diagnosis of benign nevus at a special site. We performed a detailed microscopic examination of all benign nevi in our cohort to determine the frequency of various characteristics at each special site. Sections were prepared from the formalin-fixed and paraffin-embedded tissue blocks and stained with a commercial PRAME antibody (#219650 [Abcam] at a 1:50 dilution) and counterstain. A trained dermatopathologist (S.S.R.) examined the stained sections and recorded the percentage of tumor cells with nuclear PRAME staining. We reported our results using previously established criteria for scoring PRAME immunohistochemistry⁷: 0 for no expression, 1+ for 1% to 25% expression, 2+ for 26% to 50% expression, 3+ for 51% to 75% expression, and 4+ for diffuse or 76% to 100% expression. Only strong clonal expression within a population of cells was graded.

Data handling and statistical testing were performed using the R Project for Statistical Computing (https://www.r-project.org/). Significance testing was performed using the Fisher exact test. Plot construction was performed using ggplot2 (https://ggplot2.tidyverse.org/).

Our study cohort included 36 special-site nevi, and the control cohort comprised 25 melanoma in situ (MIS) or invasive melanoma (IM) lesions occurring at special sites. Table 1 provides a breakdown of the study and control cohorts by lesion site. Table 2 details the results of our microscopic examination, describing frequency of various characteristics of special-site nevi stratified by site.

Of the 36 special-site nevi in our cohort, 20 (56%) had no staining (0) for PRAME, 11 (31%) demonstrated 1+ PRAME expression, 3 (8%) demonstrated 2+ PRAME expression, and 2 (6%) demonstrated 3+ PRAME expression. No nevi showed 4+ expression. In the control cohort, 24 of 25 (96%) MIS and IM showed 3+ or 4+ expression, with 21 (84%) demonstrating ­diffuse/4+ expression. One control case (4%) demonstrated 0 PRAME expression. These data are summarized in Table 3 and Figure 1. There is a significant difference in diffuse (4+) PRAME expression between special-site nevi and MIS/IM occurring at special sites (P=1.039×10^-12).

Based on our cohort, a positivity threshold of 3+ for PRAME expression for the diagnosis of melanoma in a special-site lesion would have a sensitivity of 96% and a specificity of 94%, while a positivity threshold of 4+ for PRAME expression would have a sensitivity of 84% and a specificity of 100%. Figures 2 through 4 show photomicrographs of a special-site nevus of the breast, which appropriately does not stain for PRAME; Figures 5 and 6 show an MIS at a special site that appropriately stains for PRAME.

Comment

The distinction between benign and malignant pigmented lesions at special sites presents a fair challenge for pathologists due to the larger degree of leniency for architectural distortion and cytologic atypia in benign lesions at these sites. The presence of architectural distortion or cytologic atypia at the lesion’s edge makes rendering a benign diagnosis especially difficult, and the need for a validated immunohistochemical stain is apparent. In our cohort, strong clonal PRAME expression provided a reliable immunohistochemical marker, allowing for the distinction of malignant lesions from benign nevi at special sites. Diffuse faint PRAME expression was present in several benign nevi within our cohort, and these lesions were considered negative (0) in our analysis.

Given the described test characteristics, we support the implementation of PRAME immunohistochemistry with a positivity threshold of 4+ expression as an ancillary test supporting the diagnosis of IM or MIS in special sites, which would allow clinicians to leverage the high specificity of 4+ PRAME expression to distinguish an IM or MIS from a benign nevus occurring at a special site. We do not recommend the use of 4+ PRAME expression as a screening test for melanoma or MIS among special-site nevi due to its comparatively low sensitivity; however, no one marker is always reliable, and we recommend continued clinicopathologic correlation for all cases. Although PRAME can assist in the delineation of malignant lesions from benign ones, microscopic examination of hematoxylin and eosin–stained section remains the gold standard for diagnosing malignant melanoma and MIS.

Although our case series included nevi and MIS/IM from all special sites, we were limited in the number of acrogenital and ear nevi included due to a relative paucity of biopsied benign nevi from these locations at the University of Virginia. Additionally, although the magnitude of the difference in PRAME expression between the study and control groups is sufficient to demonstrate statistical significance, the overall strength of our argument would be increased with a larger study group. We were limited by the number of cases available at our institution, which did not utilize PRAME during the initial diagnosis of the case; including these cases in the study group would have undermined the integrity of our argument because the differentiation of benign vs malignant initially was made using PRAME immunohistochemistry.

Conclusion

Due to their atypical features, special-site nevi can be challenging to assess. In this study, we showed that PRAME expression can be a reliable marker to distinguish benign from malignant lesions. Our results showed that 100% of benign special-site nevi demonstrated 3+ expression or less, with 56% (20/36) demonstrating no expression at all. The presence of diffuse PRAME expression (4+ PRAME staining) appears to be a specific indicator of a malignant lesion, but results should always be interpreted with respect to the patient’s clinical history and the lesion’s histomorphologic features. Further study of a larger sample size would allow refinement of the sensitivity and specificity of diffuse PRAME expression in the determination of malignancy for special-site lesions.

Acknowledgment—The authors thank the pathologistsat the University of Virginia Biorepository and Tissue Research Facility (Charlottesville, Virginia) for their skill and expertise in performing immunohistochemical staining for this study.

The assessment and diagnosis of melanocytic lesions can present a formidable challenge to even a seasoned pathologist, which is especially true when dealing with the subset of nevi occurring at special sites—where baseline variations inherent to particular locations on the body can preclude the use of features routinely used to diagnose malignancy elsewhere. These so-called special-site nevi previously have been described in the literature along with suggested criteria for differentiating malignant lesions from their benign counterparts.¹ Locations generally considered to be special sites include the acral skin, anogenital region, breast, ear, and flexural regions.^1,2

When evaluating non–special-site melanocytic lesions, general characteristics associated with a malignant diagnosis include confluence or pagetoid spread of melanocytes, nuclear pleomorphism, cytologic atypia, and irregular architecture³; however, these features can be compatible with a benign diagnosis in special-site nevi depending on their extent and the site in question. Although they can be atypical, special-site nevi tend to have the bulk of their architectural distortion and cytologic atypia in the center of the lesion as opposed to the edges.¹ If a given lesion is from a special site but lacks this reassuring feature, special care should be taken to rule out malignancy.

Preferentially expressed antigen in melanoma (PRAME) is an antigen first identified in tumor-reactive T-cell populations in patients with malignant melanoma. It is the product of an oncogene that frequently is overexpressed in melanomas, lung squamous cell carcinomas, sarcomas, and acute leukemias.⁴ It functions as an antagonist of the retinoic acid signaling pathway, which normally serves to induce further cell differentiation, senescence, or apoptosis.⁵ PRAME inhibits retinoid signaling by forming a complex with both the ligand-bound retinoic acid holoreceptor and the polycomb protein EZH2, which blocks retinoid-dependent gene expression by encouraging chromatin condensation at the RARβ promoter site⁵; therefore, expressing PRAME allows lesional cells a substantial growth advantage.

PRAME expression has been extensively characterized in non–special-site nevi and has filled the need for a rather specific marker of melanoma.^6-10 Although PRAME has been studied in acral nevi,¹¹ the expression pattern in nevi of special sites has yet to be elucidated. Herein, we present a dataset characterizing PRAME expression in these challenging lesions.

Methods

We performed a retrospective case review at the University of Virginia (Charlottesville, Virginia) and collected a panel of 36 special-site nevi that previously were diagnosed as benign by a trained dermatopathologist from January 2020 through December 2022. Special-site nevi were identified using a natural language filter for the following terms: acral, palm, sole, ear, auricular, lip, axilla, armpit, breast, groin, labia, vulva, umbilicus, and penis. This study was approved by the University of Virginia institutional review board.

The original hematoxylin and eosin slides used for primary diagnosis were re-examined to verify the prior diagnosis of benign nevus at a special site. We performed a detailed microscopic examination of all benign nevi in our cohort to determine the frequency of various characteristics at each special site. Sections were prepared from the formalin-fixed and paraffin-embedded tissue blocks and stained with a commercial PRAME antibody (#219650 [Abcam] at a 1:50 dilution) and counterstain. A trained dermatopathologist (S.S.R.) examined the stained sections and recorded the percentage of tumor cells with nuclear PRAME staining. We reported our results using previously established criteria for scoring PRAME immunohistochemistry⁷: 0 for no expression, 1+ for 1% to 25% expression, 2+ for 26% to 50% expression, 3+ for 51% to 75% expression, and 4+ for diffuse or 76% to 100% expression. Only strong clonal expression within a population of cells was graded.

Data handling and statistical testing were performed using the R Project for Statistical Computing (https://www.r-project.org/). Significance testing was performed using the Fisher exact test. Plot construction was performed using ggplot2 (https://ggplot2.tidyverse.org/).

Our study cohort included 36 special-site nevi, and the control cohort comprised 25 melanoma in situ (MIS) or invasive melanoma (IM) lesions occurring at special sites. Table 1 provides a breakdown of the study and control cohorts by lesion site. Table 2 details the results of our microscopic examination, describing frequency of various characteristics of special-site nevi stratified by site.

Of the 36 special-site nevi in our cohort, 20 (56%) had no staining (0) for PRAME, 11 (31%) demonstrated 1+ PRAME expression, 3 (8%) demonstrated 2+ PRAME expression, and 2 (6%) demonstrated 3+ PRAME expression. No nevi showed 4+ expression. In the control cohort, 24 of 25 (96%) MIS and IM showed 3+ or 4+ expression, with 21 (84%) demonstrating ­diffuse/4+ expression. One control case (4%) demonstrated 0 PRAME expression. These data are summarized in Table 3 and Figure 1. There is a significant difference in diffuse (4+) PRAME expression between special-site nevi and MIS/IM occurring at special sites (P=1.039×10^-12).

Based on our cohort, a positivity threshold of 3+ for PRAME expression for the diagnosis of melanoma in a special-site lesion would have a sensitivity of 96% and a specificity of 94%, while a positivity threshold of 4+ for PRAME expression would have a sensitivity of 84% and a specificity of 100%. Figures 2 through 4 show photomicrographs of a special-site nevus of the breast, which appropriately does not stain for PRAME; Figures 5 and 6 show an MIS at a special site that appropriately stains for PRAME.

Comment

The distinction between benign and malignant pigmented lesions at special sites presents a fair challenge for pathologists due to the larger degree of leniency for architectural distortion and cytologic atypia in benign lesions at these sites. The presence of architectural distortion or cytologic atypia at the lesion’s edge makes rendering a benign diagnosis especially difficult, and the need for a validated immunohistochemical stain is apparent. In our cohort, strong clonal PRAME expression provided a reliable immunohistochemical marker, allowing for the distinction of malignant lesions from benign nevi at special sites. Diffuse faint PRAME expression was present in several benign nevi within our cohort, and these lesions were considered negative (0) in our analysis.

Given the described test characteristics, we support the implementation of PRAME immunohistochemistry with a positivity threshold of 4+ expression as an ancillary test supporting the diagnosis of IM or MIS in special sites, which would allow clinicians to leverage the high specificity of 4+ PRAME expression to distinguish an IM or MIS from a benign nevus occurring at a special site. We do not recommend the use of 4+ PRAME expression as a screening test for melanoma or MIS among special-site nevi due to its comparatively low sensitivity; however, no one marker is always reliable, and we recommend continued clinicopathologic correlation for all cases. Although PRAME can assist in the delineation of malignant lesions from benign ones, microscopic examination of hematoxylin and eosin–stained section remains the gold standard for diagnosing malignant melanoma and MIS.

Although our case series included nevi and MIS/IM from all special sites, we were limited in the number of acrogenital and ear nevi included due to a relative paucity of biopsied benign nevi from these locations at the University of Virginia. Additionally, although the magnitude of the difference in PRAME expression between the study and control groups is sufficient to demonstrate statistical significance, the overall strength of our argument would be increased with a larger study group. We were limited by the number of cases available at our institution, which did not utilize PRAME during the initial diagnosis of the case; including these cases in the study group would have undermined the integrity of our argument because the differentiation of benign vs malignant initially was made using PRAME immunohistochemistry.

Conclusion

Due to their atypical features, special-site nevi can be challenging to assess. In this study, we showed that PRAME expression can be a reliable marker to distinguish benign from malignant lesions. Our results showed that 100% of benign special-site nevi demonstrated 3+ expression or less, with 56% (20/36) demonstrating no expression at all. The presence of diffuse PRAME expression (4+ PRAME staining) appears to be a specific indicator of a malignant lesion, but results should always be interpreted with respect to the patient’s clinical history and the lesion’s histomorphologic features. Further study of a larger sample size would allow refinement of the sensitivity and specificity of diffuse PRAME expression in the determination of malignancy for special-site lesions.

Acknowledgment—The authors thank the pathologistsat the University of Virginia Biorepository and Tissue Research Facility (Charlottesville, Virginia) for their skill and expertise in performing immunohistochemical staining for this study.