Tumor markers may predict anti–PD-L1 treatment response

NATIONAL HARBOR, MD. – Density in tumors of two key immune cells may be a biomarker for response to an inhibitor of programmed death–ligand 1 (PD-L1) among patients with non–small cell lung cancer (NSCLC).

Patients with NSCLC whose tumors bore high densities of both CD8-positive cytotoxic T lymphocytes and PD-L1 had significantly better outcomes when treated with the investigational PD-L1 inhibitor durvalumab than patients with high densities of either cell type alone, reported Sonja Althammer, PhD, a scientist at Definiens AG, maker of the anti–PD-L1 compound.

“We were looking at the predictive character of PD-L1 and CD8, and interestingly, we found out that it is a combination of the two that best predicts the response to durvalumab of the patients,” she said at the Society for Immunotherapy of Cancer annual meeting.

Dr. Althammer and her colleagues assessed the use of an automated image analysis and pattern recognition system to determine whether tumor-infiltrating CD8-positive cytotoxic T lymphocytes and PD-L1 densities could identify patients most likely to respond to the investigational PD-L1 inhibitor durvalumab.

“The hypothesis that we had in mind was that the interaction between both cell populations is important, so both cell populations should be present,” she said.

The investigators analyzed archived or fresh tumor biopsy samples from 163 patients with untreated or previously treated NSCLC (median 3 prior lines of therapy) who were enrolled in a nonrandomized phase I/II trial evaluating durvalumab in advanced NSCLC and other solid tumors.

They matched CD8 and PD-L1 immunohistochemistry-stained sections from tissues blocks. High PD-L1 expression was defined as a 25% or higher proportion of PD-L1–positive tumor cells with membrane staining at any intensity.

The images were then evaluated with an automated system, with results matched to clinical outcomes based on the densities of CD8-positive and PD-L1–positive cells, and PD-L1 alone from pre-treatment biopsy samples using a discovery set (84 samples) and validation set (79). The datasets were matched on baseline PD-L1 status, histology, Eastern Cooperative Oncology Group performance status, lines of therapy, and response.

In a training set looking at overall response rate (ORR), they found evidence to suggest that the combination of CD8 and PD-L1 positivity was associated with a higher ORR: 42%, compared with 31% for CD8-positive expression alone, 27% for PD-L1–positive alone, and 7% for CD8-positive PD-L1–negative (less than 25% expression) alone.

They then examined the predictive ability of the combined markers, and found that over approximately 28 months of follow-up, patients with CD8-positive and PD-L1–positive dense tumors had better overall survival (median overall survival, 24.3 months), compared with PD-L1–positive only patients (median overall survival, 17.1 months), CD8-positive only patients (median overall survival, 17.8 months), or the entire patient sample (median overall survival, 11.1 months).

Similarly, progression-free survival was also better among patients with high expression of both markers (respective median progression-free survival was 7.3, 3.6, 5.3, and 2.8 months).

Dr. Althammer acknowledged that the findings were preliminary and needed to be confirmed independently in larger studies.

The study was supported by MedImmune. Sonja Althammer is an employee of Definiens AG, which is a subsidiary of MedImmune/AstraZeneca.

NATIONAL HARBOR, MD. – Density in tumors of two key immune cells may be a biomarker for response to an inhibitor of programmed death–ligand 1 (PD-L1) among patients with non–small cell lung cancer (NSCLC).

Patients with NSCLC whose tumors bore high densities of both CD8-positive cytotoxic T lymphocytes and PD-L1 had significantly better outcomes when treated with the investigational PD-L1 inhibitor durvalumab than patients with high densities of either cell type alone, reported Sonja Althammer, PhD, a scientist at Definiens AG, maker of the anti–PD-L1 compound.

“We were looking at the predictive character of PD-L1 and CD8, and interestingly, we found out that it is a combination of the two that best predicts the response to durvalumab of the patients,” she said at the Society for Immunotherapy of Cancer annual meeting.

Dr. Althammer and her colleagues assessed the use of an automated image analysis and pattern recognition system to determine whether tumor-infiltrating CD8-positive cytotoxic T lymphocytes and PD-L1 densities could identify patients most likely to respond to the investigational PD-L1 inhibitor durvalumab.

“The hypothesis that we had in mind was that the interaction between both cell populations is important, so both cell populations should be present,” she said.

The investigators analyzed archived or fresh tumor biopsy samples from 163 patients with untreated or previously treated NSCLC (median 3 prior lines of therapy) who were enrolled in a nonrandomized phase I/II trial evaluating durvalumab in advanced NSCLC and other solid tumors.

They matched CD8 and PD-L1 immunohistochemistry-stained sections from tissues blocks. High PD-L1 expression was defined as a 25% or higher proportion of PD-L1–positive tumor cells with membrane staining at any intensity.

The images were then evaluated with an automated system, with results matched to clinical outcomes based on the densities of CD8-positive and PD-L1–positive cells, and PD-L1 alone from pre-treatment biopsy samples using a discovery set (84 samples) and validation set (79). The datasets were matched on baseline PD-L1 status, histology, Eastern Cooperative Oncology Group performance status, lines of therapy, and response.

In a training set looking at overall response rate (ORR), they found evidence to suggest that the combination of CD8 and PD-L1 positivity was associated with a higher ORR: 42%, compared with 31% for CD8-positive expression alone, 27% for PD-L1–positive alone, and 7% for CD8-positive PD-L1–negative (less than 25% expression) alone.

They then examined the predictive ability of the combined markers, and found that over approximately 28 months of follow-up, patients with CD8-positive and PD-L1–positive dense tumors had better overall survival (median overall survival, 24.3 months), compared with PD-L1–positive only patients (median overall survival, 17.1 months), CD8-positive only patients (median overall survival, 17.8 months), or the entire patient sample (median overall survival, 11.1 months).

Similarly, progression-free survival was also better among patients with high expression of both markers (respective median progression-free survival was 7.3, 3.6, 5.3, and 2.8 months).

Dr. Althammer acknowledged that the findings were preliminary and needed to be confirmed independently in larger studies.

The study was supported by MedImmune. Sonja Althammer is an employee of Definiens AG, which is a subsidiary of MedImmune/AstraZeneca.

NATIONAL HARBOR, MD. – Density in tumors of two key immune cells may be a biomarker for response to an inhibitor of programmed death–ligand 1 (PD-L1) among patients with non–small cell lung cancer (NSCLC).

Patients with NSCLC whose tumors bore high densities of both CD8-positive cytotoxic T lymphocytes and PD-L1 had significantly better outcomes when treated with the investigational PD-L1 inhibitor durvalumab than patients with high densities of either cell type alone, reported Sonja Althammer, PhD, a scientist at Definiens AG, maker of the anti–PD-L1 compound.

“We were looking at the predictive character of PD-L1 and CD8, and interestingly, we found out that it is a combination of the two that best predicts the response to durvalumab of the patients,” she said at the Society for Immunotherapy of Cancer annual meeting.

Dr. Althammer and her colleagues assessed the use of an automated image analysis and pattern recognition system to determine whether tumor-infiltrating CD8-positive cytotoxic T lymphocytes and PD-L1 densities could identify patients most likely to respond to the investigational PD-L1 inhibitor durvalumab.

“The hypothesis that we had in mind was that the interaction between both cell populations is important, so both cell populations should be present,” she said.

The investigators analyzed archived or fresh tumor biopsy samples from 163 patients with untreated or previously treated NSCLC (median 3 prior lines of therapy) who were enrolled in a nonrandomized phase I/II trial evaluating durvalumab in advanced NSCLC and other solid tumors.

They matched CD8 and PD-L1 immunohistochemistry-stained sections from tissues blocks. High PD-L1 expression was defined as a 25% or higher proportion of PD-L1–positive tumor cells with membrane staining at any intensity.

The images were then evaluated with an automated system, with results matched to clinical outcomes based on the densities of CD8-positive and PD-L1–positive cells, and PD-L1 alone from pre-treatment biopsy samples using a discovery set (84 samples) and validation set (79). The datasets were matched on baseline PD-L1 status, histology, Eastern Cooperative Oncology Group performance status, lines of therapy, and response.

In a training set looking at overall response rate (ORR), they found evidence to suggest that the combination of CD8 and PD-L1 positivity was associated with a higher ORR: 42%, compared with 31% for CD8-positive expression alone, 27% for PD-L1–positive alone, and 7% for CD8-positive PD-L1–negative (less than 25% expression) alone.

They then examined the predictive ability of the combined markers, and found that over approximately 28 months of follow-up, patients with CD8-positive and PD-L1–positive dense tumors had better overall survival (median overall survival, 24.3 months), compared with PD-L1–positive only patients (median overall survival, 17.1 months), CD8-positive only patients (median overall survival, 17.8 months), or the entire patient sample (median overall survival, 11.1 months).

Similarly, progression-free survival was also better among patients with high expression of both markers (respective median progression-free survival was 7.3, 3.6, 5.3, and 2.8 months).

Dr. Althammer acknowledged that the findings were preliminary and needed to be confirmed independently in larger studies.

The study was supported by MedImmune. Sonja Althammer is an employee of Definiens AG, which is a subsidiary of MedImmune/AstraZeneca.