In the Journals

Pembrolizumab induces durable responses in relapsed, refractory primary mediastinal large B-cell lymphoma

Philippe Armand
Philippe Armand

Pembrolizumab appeared to be associated with high response rates, durable activity and manageable toxicity among patients with relapsed or refractory primary mediastinal large B-cell lymphoma, according to results of the phase 1b KEYNOTE-013 and phase 2 KETNOTE-170 trials published jointly in Journal of Clinical Oncology.

“The standard of care for relapsed/refractory primary mediastinal large B-cell lymphoma remains salvage and [autologous stem cell] transplant for first relapse, but following that there was no standard because most therapies are ineffective,” Philippe Armand, MD, PhD, chief of the division of lymphoma and director of clinical research of the lymphoma program at Dana-Farber Cancer Institute, told HemOnc Today. “Recently, CAR T cells have been approved in this setting and provide a valuable option for patients, although we do not well-understand their exact efficacy in this disease.”

Genetic aberrations at 9p24.1 in primary mediastinal large B-cell lymphoma result in overexpression of PD-L1 and PD-L2, suggesting the disease may be susceptible to PD-1 blockade.

KEYNOTE-013

In the single-arm, phase 1B KEYNOTE-013 trial, Armand and colleagues assigned 21 patients (median age, 31 years; range, 22-62; 67% women) with relapsed or refractory primary mediastinal large B-cell lymphoma to pembrolizumab (Keytruda, Merck), a humanized IgG4 monoclonal antibody that targets PD-1.

Patients received a median three (range, 2-9) prior lines of therapy, including rituximab (Rituxan, Genentech). Fifteen received prior radiation, and eight underwent prior stem cell transplantation.

The first 10 patients received 10 mg/kg pembrolizumab every 2 weeks. The other 11 patients received 200 mg every 3 weeks starting on day 1 of each treatment cycle. Treatment continued for a maximum 35 cycles or 2 years until disease progression, unacceptable toxicity or patient withdrawal.

Objective response by investigator assessment and safety served as the primary endpoints. Duration of response, PFS and OS served as secondary endpoints. Associations between biomarkers and pembrolizumab activity served as exploratory endpoints.

Median follow-up was 29.1 months (range, 0.6-49.6).

Results showed an objective response rate of 48% (95% CI, 26-70), with seven patients achieving complete response.

Median time to response was 2.7 months (range, 1.4-3.4), and median duration was not reached (range, 1.9+ months to 39.8+ months).

Thirteen patients experienced a reduction in target lesion size from baseline, and two patients converted from partial response to complete response after about 1 year.

Median PFS was 10.4 months (95% CI, 3.4-not reached), with an estimated 12-month PFS rate of 47%.

Eleven patients had died by data cutoff. Median OS was 31.4 months (95% CI, 4.9-not reached).

Five patients (24%) experienced grade 3 or grade 4 treatment-related adverse events, with one discontinuing treatment because of febrile neutropenia.

KEYNOTE-170

In the single-arm, phase 2 KEYNOTE-170 trial, researchers assigned 53 patients (median age, 33 years; range, 20-61; 57% women) with relapsed or refractory primary mediastinal large B-cell lymphoma to 200 mg pembrolizumab every 3 weeks for up to 2 years or until disease progression, unacceptable toxicity or patients withdrawal.

Patients received a median three (range, 2-8) prior lines of therapy, including rituximab. Seventeen patients had also received radiation and 14 underwent prior stem cell transplantation.

Objective response by central review served as the primary endpoint. Duration of response, PFS, OS, safety and tolerability served as secondary endpoints. Associations between biomarkers and pembrolizumab activity served as exploratory endpoints.

Median follow-up was 12.5 months (range, 0.1-25.6).

Results showed an ORR of 45% (95% CI, 32-60), with seven patients achieving complete response.

Median time to response was 2.9 months (range, 2.4-8.5), and median duration was not reached (range, 1.1+ months to 22+ months).

Thirty patients experienced a reduction in target lesion size from baseline, and one patient converted from partial response to complete response after about 1 year.

Median PFS was 5.5 months (95% CI, 2.8-12.1), with an estimated 12-month PFS rate of 38%.

Twenty-four patients had died by data cutoff. Median OS was not reached (95% CI, 7.3-not reached).

Twelve patients (23%) experienced grade 3 or grade 4 treatment-related adverse events, with one discontinuing treatment because of increased aspartate aminotransferase. Three patients died of adverse events not related to treatment.

Biomarker analyses

Biomarker analyses of 42 evaluable samples from patients in both studies showed an association between the extent of 9p24 gene abnormality and PD-L1 expression, which in turn appeared significantly associated with PFS.

In 40 samples evaluable for 9p24.1 status by fluorescence in situ hybridization, 9p24.1 ranged from single copy loss (n = 1) to disomy (n = 1), polysomy (n = 9), copy gain (n = 19), amplification (n = 6) or rearrangements (n = 4) involving PD-L1/ PD-L2, according to the study.

Researchers observed an increase in PD-L1 expression with increasing levels of PD-L1/PD-L2 DNA copy number gain.

Objective response rates increased with PD-L1 H-score, from 25% among patients with an H-score of zero to 42% for those with an H-score of 1 to 99 and 64% with an H-score of 100 or greater.

“I hope that PD-1 blockade can be moved to the front-line setting in combination with chemotherapy, although of course this will require clinical trials,” Armand said. “I do hope that eventually randomized clinical trials in one of those settings can document an improvement in OS with PD-1 blockade.” – by John DeRosier

For more information:

Philippe Armand, MD, PhD, can be reached at Dana-Farber Cancer Institute, 450 Brookline Ave., Boston, MA 02215; email: philippe_armand@dfci.harvard.edu.

Disclosures: Merck Sharp & Dohme funded this study. Armand reports honoraria, research funding and travel expenses from, as well as consultant/advisory roles with, Bristol-Myers Squibb and Merck Sharp & Dohme; consultant roles with and research funding from Adaptive Biotechnologies and Affimed Therapeutics; research funding from Genentech, IGM and Tensha Therapeutics; and travel expenses from Sequenta. Please see the study for all other authors’ relevant financial disclosures.

Philippe Armand
Philippe Armand

Pembrolizumab appeared to be associated with high response rates, durable activity and manageable toxicity among patients with relapsed or refractory primary mediastinal large B-cell lymphoma, according to results of the phase 1b KEYNOTE-013 and phase 2 KETNOTE-170 trials published jointly in Journal of Clinical Oncology.

“The standard of care for relapsed/refractory primary mediastinal large B-cell lymphoma remains salvage and [autologous stem cell] transplant for first relapse, but following that there was no standard because most therapies are ineffective,” Philippe Armand, MD, PhD, chief of the division of lymphoma and director of clinical research of the lymphoma program at Dana-Farber Cancer Institute, told HemOnc Today. “Recently, CAR T cells have been approved in this setting and provide a valuable option for patients, although we do not well-understand their exact efficacy in this disease.”

Genetic aberrations at 9p24.1 in primary mediastinal large B-cell lymphoma result in overexpression of PD-L1 and PD-L2, suggesting the disease may be susceptible to PD-1 blockade.

KEYNOTE-013

In the single-arm, phase 1B KEYNOTE-013 trial, Armand and colleagues assigned 21 patients (median age, 31 years; range, 22-62; 67% women) with relapsed or refractory primary mediastinal large B-cell lymphoma to pembrolizumab (Keytruda, Merck), a humanized IgG4 monoclonal antibody that targets PD-1.

Patients received a median three (range, 2-9) prior lines of therapy, including rituximab (Rituxan, Genentech). Fifteen received prior radiation, and eight underwent prior stem cell transplantation.

The first 10 patients received 10 mg/kg pembrolizumab every 2 weeks. The other 11 patients received 200 mg every 3 weeks starting on day 1 of each treatment cycle. Treatment continued for a maximum 35 cycles or 2 years until disease progression, unacceptable toxicity or patient withdrawal.

Objective response by investigator assessment and safety served as the primary endpoints. Duration of response, PFS and OS served as secondary endpoints. Associations between biomarkers and pembrolizumab activity served as exploratory endpoints.

Median follow-up was 29.1 months (range, 0.6-49.6).

Results showed an objective response rate of 48% (95% CI, 26-70), with seven patients achieving complete response.

Median time to response was 2.7 months (range, 1.4-3.4), and median duration was not reached (range, 1.9+ months to 39.8+ months).

Thirteen patients experienced a reduction in target lesion size from baseline, and two patients converted from partial response to complete response after about 1 year.

Median PFS was 10.4 months (95% CI, 3.4-not reached), with an estimated 12-month PFS rate of 47%.

PAGE BREAK

Eleven patients had died by data cutoff. Median OS was 31.4 months (95% CI, 4.9-not reached).

Five patients (24%) experienced grade 3 or grade 4 treatment-related adverse events, with one discontinuing treatment because of febrile neutropenia.

KEYNOTE-170

In the single-arm, phase 2 KEYNOTE-170 trial, researchers assigned 53 patients (median age, 33 years; range, 20-61; 57% women) with relapsed or refractory primary mediastinal large B-cell lymphoma to 200 mg pembrolizumab every 3 weeks for up to 2 years or until disease progression, unacceptable toxicity or patients withdrawal.

Patients received a median three (range, 2-8) prior lines of therapy, including rituximab. Seventeen patients had also received radiation and 14 underwent prior stem cell transplantation.

Objective response by central review served as the primary endpoint. Duration of response, PFS, OS, safety and tolerability served as secondary endpoints. Associations between biomarkers and pembrolizumab activity served as exploratory endpoints.

Median follow-up was 12.5 months (range, 0.1-25.6).

Results showed an ORR of 45% (95% CI, 32-60), with seven patients achieving complete response.

Median time to response was 2.9 months (range, 2.4-8.5), and median duration was not reached (range, 1.1+ months to 22+ months).

Thirty patients experienced a reduction in target lesion size from baseline, and one patient converted from partial response to complete response after about 1 year.

Median PFS was 5.5 months (95% CI, 2.8-12.1), with an estimated 12-month PFS rate of 38%.

Twenty-four patients had died by data cutoff. Median OS was not reached (95% CI, 7.3-not reached).

Twelve patients (23%) experienced grade 3 or grade 4 treatment-related adverse events, with one discontinuing treatment because of increased aspartate aminotransferase. Three patients died of adverse events not related to treatment.

Biomarker analyses

Biomarker analyses of 42 evaluable samples from patients in both studies showed an association between the extent of 9p24 gene abnormality and PD-L1 expression, which in turn appeared significantly associated with PFS.

In 40 samples evaluable for 9p24.1 status by fluorescence in situ hybridization, 9p24.1 ranged from single copy loss (n = 1) to disomy (n = 1), polysomy (n = 9), copy gain (n = 19), amplification (n = 6) or rearrangements (n = 4) involving PD-L1/ PD-L2, according to the study.

Researchers observed an increase in PD-L1 expression with increasing levels of PD-L1/PD-L2 DNA copy number gain.

Objective response rates increased with PD-L1 H-score, from 25% among patients with an H-score of zero to 42% for those with an H-score of 1 to 99 and 64% with an H-score of 100 or greater.

PAGE BREAK

“I hope that PD-1 blockade can be moved to the front-line setting in combination with chemotherapy, although of course this will require clinical trials,” Armand said. “I do hope that eventually randomized clinical trials in one of those settings can document an improvement in OS with PD-1 blockade.” – by John DeRosier

For more information:

Philippe Armand, MD, PhD, can be reached at Dana-Farber Cancer Institute, 450 Brookline Ave., Boston, MA 02215; email: philippe_armand@dfci.harvard.edu.

Disclosures: Merck Sharp & Dohme funded this study. Armand reports honoraria, research funding and travel expenses from, as well as consultant/advisory roles with, Bristol-Myers Squibb and Merck Sharp & Dohme; consultant roles with and research funding from Adaptive Biotechnologies and Affimed Therapeutics; research funding from Genentech, IGM and Tensha Therapeutics; and travel expenses from Sequenta. Please see the study for all other authors’ relevant financial disclosures.

    See more from Immuno-Oncology Resource Center