The Journal of Community and Supportive Oncology

This article reviews the mechanism of action, clinical trial results, and adverse effects of two molecularly targeted anticancer agents, the tyrosine kinase inhibitors (TKIs) sorafenib (Nexavar^®; Bayer HealthCare Pharmaceuticals, Montville, NJ, and Onyx Pharmaceuticals, Emeryville, CA) and sunitinib (Sutent^®; Pfizer Pharmaceuticals, New York, NY). This article specifically focuses on the diagnosis and management of TKI-associated hand–foot skin reaction (HFSR) from the perspectives of the medical oncologist, the dermatologist, and the oncology nurse. Data were derived from (1) published reports of preclinical and phase I–III clinical studies of sorafenib and sunitinib and (2) published abstracts of phase II–III clinical trials of sorafenib and sunitinib.

The Medical Oncologist's Viewpoint

Molecularly Targeted Agents

Molecularly targeted therapies are directed at specific mechanisms involved in cell division, invasion, and metastasis, as well as in cell survival mediated by avoidance of apoptosis and resistance to conventional treatments. Clinical trials in several cancer types have shown that these TKIs can inhibit these activities of cancer cells by either cytostatic or cytotoxic mechanisms.¹ However, the ability of these agents to inhibit multiple cancer cell pathways via novel mechanisms of action may explain, at least in part, their apparent direct toxic effects.² These include adverse events that, from a medical viewpoint, must be anticipated, promptly recognized, and properly treated. Doing so can help minimize disruption to the patient's quality of life and may reduce the need for dose reduction or treatment interruption.¹

Both sorafenib and sunitinib are orally administered, small-molecule inhibitors of multiple kinases, some of which are common to both agents (Figure 2).³ Sorafenib has known effects on tumor-cell proliferation and angiogenesis. Its antiproliferative effects are exerted via inhibition of serine/threonine kinases of the RAF/MEK/ERK signaling pathway (also called the MAP-kinase pathway) that is found within tumor cells; specifically, sorafenib targets wild-type RAF gene products (CRAF, BRAF) and mutant BRAF. The antiangiogenic effects of sorafenib are exerted via its inhibition of extracellular vascular endothelial growth factor (VEGF) receptors 2 and 3 (VEGFR-2 and VEGFR-3) and platelet-derived growth factor receptor beta (PDGFR-β), which is found mainly in the tumor vasculature. Sorafenib also exerts broad-spectrum activity against the stem-cell growth factor receptor (c-KIT), FMS-like tyrosine kinase 3 (Flt3), and the receptor encoded by the ret proto-oncogene (RET).^{[4], [5], [6] and [7]} Sunitinib has demonstrated effects on the growth, pathologic angiogenesis, and metastatic progression of cancer by inhibiting PDGFR-α and -β; VEGFR-1, -2, and -3; and colony-stimulating factor receptor (CSF-1R), c-KIT, Flt3, and RET.⁸

Sorafenib was approved for the treatment of advanced renal cell carcinoma (RCC) in 2005 and for unresectable hepatocellular carcinoma (HCC) in 2007. The efficacy of sorafenib in 903 patients with advanced RCC was demonstrated in the phase III Treatment Approaches in Renal Cancer Global Evaluation Trial (TARGET), the largest phase III trial ever conducted in the second-line setting in patients with advanced RCC. Sorafenib significantly enhanced median progression-free survival (PFS) compared with placebo (24 vs 12 weeks),⁹ which led to early termination of the study and crossover from placebo to active drug. A preplanned analysis, which did not include patients who received placebo (who had crossed over to active treatment), ultimately demonstrated that sorafenib significantly prolonged overall survival (OS).¹⁰ Furthermore, 84% of sorafenib-treated patients experienced a clinical benefit, defined as objective response or disease stabilization.⁹ These results have been confirmed in larger, “real-world” patient populations in expanded-access programs conducted in North America (n = 2504)¹¹ and the European Union (n = 118).¹²

Definitive data supporting the efficacy of sorafenib in HCC were provided by the randomized, double-blind, placebo-controlled Sorafenib CCC Assessment Randomized Protocol (SHARP) trial, the largest phase III trial ever conducted in patients with advanced HCC (n = 599) and the first phase III study to demonstrate a significant survival advantage with a systemic treatment in advanced HCC. In this trial, patients treated with sorafenib experienced a 44% increase in median OS (10.7 vs 7.9 months, hazard ratio [HR] = 0.69, 95% confidence interval [CI] 0.55–0.87, P < 0.001) and a 73% prolongation in median time to radiographic progression (5.5 vs 2.8 months, P < 0.001) compared with patients who received placebo.¹³ These results were confirmed in a separate phase III, randomized, double-blind, placebo-controlled study conducted in 226 patients from the Asia-Pacific region with advanced HCC.¹⁴ In this trial also, sorafenib significantly prolonged median OS (6.5 vs 4.2 months, HR = 0.68, 95% CI 0.50–0.93, P = 0.014) and time to progression (TTP) (2.8 vs 1.4 months, HR = 0.57, 95% CI 0.42–0.79, P = 0.0005) compared with placebo.

Sunitinib received approval in 2006 for use in patients with gastrointestinal stromal tumor (GIST) whose disease is refractory to imatinib (Gleevec^®; Novartis Pharmaceuticals, East Hanover, NJ) or who are intolerant to the drug and in those with advanced RCC. Approval of sunitinib for the treatment of GIST was based on data from a randomized, placebo-controlled, phase III trial of 312 patients with imatinib-refractory GIST.¹⁵ In that study, sunitinib treatment increased median PFS (24.1 vs 6.0 weeks, HR = 0.33, 95% CI 0.24–0.47, P < 0.0001) and median TTP (27.3 vs 6.4 weeks, HR = 0.33, 95% CI 0.23–0.47, P < 0.0001) compared with placebo. The trial was unblinded early when a planned interim analysis revealed significantly longer TTP with sunitinib than with placebo. A subsequent analysis showed that median OS with sunitinib was about twice that with placebo (73.9 vs 35.7 weeks, P < 0.001).¹⁶ In an ongoing, worldwide treatment-use program to provide expanded access to sunitinib for patients with advanced GIST intolerant of or resistant to imatinib,¹⁷ sunitinib treatment resulted in a median estimated TTP of 41 weeks and a median estimated OS of 75 weeks in the population analyzed (n = 1,117).

A separate phase III randomized controlled trial was conducted in 750 patients with advanced RCC and no history of systemic therapy for RCC.^{[18] and [19]} The active comparator in this trial was interferon-alfa (IFN-α). Compared with IFN-α, sunitinib significantly increased median PFS (11 vs 5 months, HR = 0.539, 95% CI 0.451–0.643, P < 0.001) and was associated with a greater objective response rate (47% vs 12%, P < 0.001). Median OS was greater in the sunitinib group (26.4 vs 21.8 months), but the difference was not significant (P = 0.051). Data from expanded-access programs in patients with RCC and GIST support the phase III trial data for sunitinib.²⁰ Efficacy data for sorafenib and sunitinib are summarized in Table 1.^{[9], [10], [11], [12], [13], [14], [15], [17], [18] and [20]}

Characteristics of Hand–Foot Skin Reaction

Data from the clinical trials for sorafenib and sunitinib indicate that both agents are generally well-tolerated; common treatment-related adverse reactions include diarrhea, alopecia, nausea, fatigue, rash, and hypertension, as well as palmar–plantar erythrodysesthesia (PPE) syndrome, also known as hand–foot skin reaction (HFSR) (Table 2).^{[10] and [19]} HFSR is a dermatologic toxicity that has been reported in 14%–62% of patients treated with sorafenib or sunitinib (Table 3).^{[9], [11], [12], [13], [14], [15], [17], [18], [20], [21], [22], [23], [24] and [25]} In general, the term HFSR refers to a group of signs and symptoms affecting the hands and feet of patients taking sorafenib, sunitinib, or, to a lesser extent, other TKIs such as pazopanib (Votrient™; GlaxoSmithKline, Research Triangle Park, NC)^{[26] and [27]} and axitinib (AG013736).^{[28], [29], [30] and [31]}

HFSR is typically characterized by redness, marked discomfort, swelling, and tingling in the palms of the hands and/or soles of the feet.³² HFSR can be painful enough to interfere profoundly with activities of daily living (ADLs). In fact, patients may report symptoms after as few as 2 weeks on TKI therapy, at which point they may present to the health-care provider (HCP) wearing slippers, unable to walk, and having difficulty in performing ADLs such as eating, dressing, and bathing.^{[1] and [33]} Although HFSR can lead to TKI dose modification or treatment discontinuation, preventive measures can be taken before TKIs are initiated to reduce the likelihood of HFSR. In addition, early treatment of symptoms may prevent HFSR from progressing to the point at which the patient's ability to receive the full potential benefit of therapy is compromised.^{[3], [34], [35] and [36]}

Signs and symptoms of HFSR may appear concomitantly or sequentially and can affect both hands and both feet. Although symptoms are most prominent on the palms and soles, other areas of the hands and feet may also be involved, including the tips of the fingers and toes, the heels, and metatarsophalangeal skin; areas of flexure; and skin overlying the metacarpophalangeal and interphalangeal joints.³ These “pressure areas” are where the most severe symptoms are typically seen. Common symptoms include dysesthesia and paresthesia, described as “tingling, prickling, or creeping sensations” and/or sensitivity or intolerance to hot or warm objects (which may occur before other symptoms are apparent); erythema; edema; hyperkeratosis; and dry and/or cracked skin.^{[1] and [34]} Actual HFSR lesions are described as tender and scaling, with a peripheral halo of erythema, yellowish and hyperkeratotic plaques, or callous-like blisters (which usually do not contain fluid), typically localized to areas of pressure.^{[3] and [35]} Desquamation, particularly with sunitinib treatment, may also be present.³⁷

Since both sorafenib and sunitinib inhibit the VEGFRs, PDGFRs, c-KIT, and Flt3,³⁸ it is likely that inhibition of one or more of these receptors and/or pathways plays a role in HFSR development.³⁶ Differences in the relative appearance of HFSR symptoms are dependent on whether sorafenib or sunitinib is used. Sunitinib use is more often associated with desquamation, whereas sorafenib is more often associated with areas of hyperkeratosis, particularly formation of thick calluses on the soles of the feet.³⁷ The timing of the first appearance of symptoms may also vary according to the TKI used. HFSR usually develops within the first 2–4 weeks of treatment with a TKI and almost always within the first 6 weeks.³⁵ However, because the severity of HFSR appears to be dose-dependent,³ signs and symptoms may present later rather than sooner in patients treated with sunitinib. This is likely due to the recommended sunitinib dosing schedule, which incorporates a 2-week period during which no drug is administered. Although HFSR frequently decreases in intensity during treatment, even without dose modifications or treatment interruption, prompt treatment of HFSR is recommended to prevent rapid progression. Early symptoms can usually be resolved easily by appropriate treatment, which often allows continuation of full-dose therapy for the prescribed length of time.

It is important to note what HFSR is not. TKI-associated HFSR is not the same clinical entity as the hand–foot syndrome (HFS) traditionally seen with cytotoxic agents such as infusional 5-fluorouracil (5-FU); capecitabine, the oral prodrug of 5-FU; and pegylated liposomal doxorubicin. Although HFSR and HFS share several clinical and pathological aspects—each previously has been called “acral erythema” and “PPE”—they clearly are not the same clinical or pathologic entity. HFSR is neither an allergic reaction to a drug nor an indication that a patient may be intolerant to a drug. Finally, HFSR does not indicate drug efficacy, as may be the case with skin rash in patients with non-small-cell lung cancer treated with erlotinib.^{[3] and [39]}

Grading HFSR

In published reports, the severity of HFSR is usually graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE),³² a three-grade classification system. However, modified criteria are considered by some to be a better fit for routine clinical practice.¹ What distinguishes the modified criteria from the NCI criteria (version 4.02) is the inclusion of HFSR-specific clinical characteristics plus certain patient-defined considerations used to categorize severity. The modified criteria expressly define the degree to which HFSR discomfort affects the patient's normal activities, an improvement over version 4.02 used alone. The NCI-CTCAE version 4.02 criteria, the modified criteria, and corresponding patient photographs are presented in Figure 3.^{[1] and [32]}

Recommendations for the treatment of grade 1 HFSR include early and appropriate dermatologic management and active collaboration among HCPs.⁴⁰

The Dermatologist's Viewpoint

Although the exact pathogenesis of HFSR has not been fully elucidated, research into its cause(s) is ongoing. Theoretically, traditional HFS is thought to be due to the direct toxic effects of drugs or their ability to invoke a “host-vs-host” response. In contrast, a unique mechanism has been proposed for TKI-associated HFSR: simultaneous blockage of VEGFRs and PDGFRs.²

Three histopathologic features have been found to predominate in HFSR: dyskeratotic keratinocytes at various states of necrosis (Figure 4⁴¹), basal layer vacuolar degeneration, and mild perivascular or lichenoid lymphocyte-predominant infiltrate.² Immunohistochemistry with a variety of skin-cell markers has shown a significant modification of normal maturation of keratinocytes, which are often apoptotic. Minor modifications of blood vessels are also seen, but no signs of intense vasculitis are evident. This is important because HFSR is suspected of being a “class effect” of TKIs that target VEGFRs. HFSR is not seen in patients treated with single-agent bevacizumab, and the lack of histologic evidence of significant damage to blood vessels suggests that HFSR does not result from the general inhibition of angiogenesis. A retrospective analysis found that HFSR rates were higher when patients were treated with sorafenib and bevacizumab in combination, supporting the hypothesis that HSFR is due to the anti-VEGF properties of sorafenib.⁴² Other possible causes of HFSR include activation by a ligand other than VEGF and/or inhibition of one of the other protein targets inhibited by both sorafenib and sunitinib.^{[3] and [35]}

Incidence and Severity of HFSR With TKI

To determine the incidence and severity of HFSR specific to sorafenib, a double-blind, prospective, dermatologic substudy was performed in patients enrolled in the phase III TARGET trial.³⁵ Eighty-five patients with RCC were randomized to receive either sorafenib (n = 43) or placebo (n = 42). Dermatologic examinations were performed before and during treatment. Ninety-one percent of sorafenib-treated patients experienced at least one cutaneous reaction compared with 7% of those in the placebo group. A variant of HFSR clinically distinct from chemotherapy-induced HFS was observed in 60% of sorafenib-treated patients. Reversible grade 3 HFSR leading to dose reduction occurred in two sorafenib-treated patients. Additional cutaneous reactions were facial erythema, scalp dysesthesia, alopecia, and subungual splinter hemorrhages.

HFSR (of any grade) has been shown to occur in approximately 30% of patients treated with sorafenib and 20% of those who received sunitinib in clinical studies.⁴³ Grade 3/4 HFSR has been observed in approximately 6% of sorafenib-treated and 5% of sunitinib-treated patients. HFSR was not reported in a phase II study of 142 patients with relapsed or refractory soft-tissue sarcoma treated with pazopanib.⁴⁴ In a phase III randomized, double-blind, placebo-controlled trial of pazopanib in patients with advanced RCC, the incidence of HFSR was <10%, while the incidence of grade 3/4 HFSR was <1%. Potential differences may be explained by variations in the potency and selectivity of the TKIs.²⁷

Management Strategies

Our work at the Dermatology Center at the Gustave-Roussy Institute has shown that early intervention against the dermatologic adverse effects of these TKIs can inhibit patient progression to a more serious form of HFSR.^{[34] and [38]}

Effective management of HFSR can begin prior to initiation of treatment with sorafenib or sunitinib. Patients should be advised to remove any preexisting hyperkeratotic areas or calluses, keep skin well-moisturized with appropriate creams, and cushion pressure points with cotton socks, soft shoes, and/or insoles. Dose modification is typically not required for grade 1 HFSR; symptomatic treatments should be employed instead.

If HFSR symptoms progress to grade 2 or 3, with pain and a decrease in quality of life, the dose of sorafenib or sunitinib can be modified until symptoms recede, after which the patient can be brought back to the full dose. Very often, the patient can tolerate the full-dose treatment simply by decreasing the dose briefly.³ A recommended dose-modification scheme is shown in Figure 5.³

TKIs are being studied in patients with additional types of tumor, possibly in the adjuvant setting, as well as in combinations. Because these drugs are administered orally, with a decreased (compared with conventional cytotoxic agents) frequency of nurse– or doctor–patient interactions, patients must be very well-informed of any potential toxicities with the TKIs.

At present, there are no evidence-based treatment guidelines for the prevention or management of HFSR. However, HCPs most involved in the day-to-day care of patients with HFSR have made great progress in establishing preventive and treatment strategies and in identifying ancillary products likely to decrease the incidence and/or severity of symptoms. Prevention, which includes preventing HFSR entirely as well as preventing progression from its initial appearance, is a key component of HFSR management.

The Oncology Nurse's Viewpoint

The nurse's viewpoint begins with patient education and empowerment. The goal is to prevent adverse effects from occurring while managing any adverse effects that do occur so that the patient has the best chance of staying on anticancer therapy. This requires a strong partnership between the HCP team and the patient. Although not all cases of HFSR can be prevented, experience suggests that symptom incidence and severity can be alleviated by educating patients to recognize the signs and symptoms of HFSR and report these to their HCPs. HFSR typically occurs early in the course of therapy, so it is prudent to be especially vigilant during the first 6 weeks. Providing the patient with a brochure about HFSR to refer to at home may facilitate identification of HFSR.

To address the lack of evidence-based guidelines to prevent or treat HFSR, an international, interdisciplinary expert panel has provided a set of consensus recommendations for the management of TKI-associated HFSR.⁴⁵ One component of these recommendations can be phrased simply for the patients as a “3C” approach to management: control calluses, comfort with cushions, and cover with creams.

Prior to treatment, the patient should receive a full-body examination, with emphasis on the condition of the hands and feet. Evaluation should be performed by a qualified HCP who can determine whether there are physical conditions that may predispose a patient to areas of increased friction or rubbing. For all patients, especially those with comorbid conditions (eg, diabetes, poor circulation), a pretreatment pedicure is highly recommended. Patients should also be educated on the proper use of tools (eg, a pumice stone) to aid in callus removal. Such tools are considered beneficial because patients can control the frequency of their use and the extent of skin removed. However, because areas of hyperkeratosis are often extremely tender and painful, patients are cautioned against overuse of these tools, including the aggressive “paring” or “cutting” of callused areas. Finally, patients should be advised of the need for clean tools to guard against infection.

Other protective measures include the use of thick cotton gloves and/or socks, which may also help the skin to retain moisture, and avoidance of warm and/or hot water or objects, tight-fitting shoes, or other items that may rub, pinch, or cause friction in affected areas. Tender areas, pressure points, and pressure-sensitive areas of the hands and feet should be protected. For example, weight lifters might be advised to wear gloves. These recommendations hold true both before and after development of HFSR.^{[3], [33] and [35]} Well-padded but nonconstrictive footwear should be worn, and the use of insole cushions or inserts (eg, silicone or gel) should be encouraged. Foot soaks with lukewarm water and magnesium sulfate may be soothing. Tender areas should be protected at all times, and patients should be encouraged not to walk barefoot.

Use of over-the-counter and prescription-strength creams and moisturizers during treatment with TKIs has also been recommended (Table 4).^{[40], [45], [46], [47], [48] and [49]} Moisturizing agents should be applied liberally, immediately after bathing. Cotton gloves and/or socks can also be worn, to help retain moisture and to provide an additional layer of protection. When applied liberally, these products soften areas of thick and hardened skin, help keep the skin pliable, and may prevent cracks or breaks in skin integrity, which could cause additional discomfort. Prescription-strength topical agents have also shown anecdotal benefit (Table 4). These topical agents are typically applied twice daily to affected areas only because they may irritate unaffected skin. Data on the use of topical/systemic corticosteroids in the treatment of HFS remain inconclusive; the literature primarily includes case studies in patients with PPE treated with chemotherapeutic agents including pegylated liposomal doxorubicin.² Finally, a qualified HCP must always be consulted to ensure proper diagnosis and treatment of HFSR.

Summary

The addition of molecularly targeted agents to anticancer treatment has been found to cause both common and novel adverse reactions. HFSR is being increasingly recognized as a potential dose-limiting toxicity associated with sorafenib or sunitinib treatment that can result in discomfort, pain, decreased quality of life, and premature termination of a potentially effective cancer treatment. It is important to educate patients about potential dermatologic adverse effects associated with TKIs because limiting toxicity can help avoid treatment interruptions or dose reductions while improving ADLs.

The precise pathogenic mechanism of HFSR is currently not known, and there is no evidence-based protocol for treatment of HFSR. However, the increased clinical experience with these agents has resulted in a wealth of published articles describing empiric and symptomatic approaches that appear to help to prevent and manage HFSR. Frequent communication is necessary between the physician and patient, particularly 2–4 weeks from the initiation of therapy. Symptoms of HFSR should be recognized as early as possible. Providing the patient with a brochure about HFSR to refer to at home may facilitate the early identification of HFSR.

Patients should be advised of the “3C” approach to the management of TKI-associated HFSR: control calluses, comfort with cushions, and cover with creams. Creams should be applied after bathing and before going to bed; cotton gloves and socks should be worn to keep the cream on the hands and feet during the night.

Symptoms of HFSR typically are manageable with the implementation of supportive measures. If symptoms worsen, dose modification or interruption will result in a return to grade 0/1. Many patients can successfully be rechallenged with the full dose. Observations across multiple viewpoints have consistently shown that HFSR severity can be reduced in patients who are educated about HFSR and proactive about its detection and management.

Acknowledgments

All authors contributed equally to the development of this report. Editorial support was provided by Katherine Wright, PharmD, RPh, ISD, Wrighter Medical Education and Training, West Hills, CA; John A. Ibelli, CMPP, BelMed Professional Resources, New Rochelle, NY; and John D. Zoidis, MD, Bayer HealthCare Pharmaceuticals, Montville, NJ.