CHICAGO -- Promising early results in metastatic triple-negative breast cancer failed to hold up in a phase III trial of the PARP inhibitor iniparib, according to data presented here.
A one-month improvement in progression-free survival with iniparib did not meet the prespecified definition of statistical significance. Adding the PARP inhibitor to chemotherapy was associated with less than a one-month improvement in overall survival compared with chemotherapy alone.
An exploratory analysis suggested that patients who received iniparib as second- or third-line therapy might have benefited from treatment, Joyce O'Shaughnessy, MD, reported at the American Society of Clinical Oncology meeting.
"Metastatic triple-negative breast cancer is a highly heterogeneous patient population on intrinsic subtyping," said O'Shaughnessy, of Baylor University Medical Center in Dallas. "Biomarker analyses are under way to evaluate patient populations that may benefit from iniparib."
PARP inhibitors block tumor cells' ability to repair DNA damage, including damage caused by chemotherapy. Preclinical studies have suggested the agents may also potentiate chemotherapy-induced DNA damage.
The drugs are being evaluated clinically in multiple tumor types. Earlier this year, O'Shaughnessy and colleagues reported final results of a phase II study showing significant improvement in progression-free survival in patients with metastatic triple-negative breast cancer treated with iniparib (N Engl J Med 2011; 364: 205-214).
However, less than a month after publication of the positive results sanofi-aventis announced that a larger phase III study had failed to corroborate the phase II results. O'Shaughnessy's presentation at ASCO confirmed the study's negative outcome.
The phase III trial involved 519 women with metastatic triple-negative breast cancer, treated with two or fewer metastatic regimens. All patients received gemcitabine and carboplatin and were randomized to iniparib or placebo. Patients who progressed on placebo could cross over to the iniparib arm, and 96% of patients in the placebo arm subsequently crossed over.
The trial had coprimary endpoints of overall survival and progression-free survival, and the study would be considered positive if either endpoint favored iniparib. Secondary endpoints included objective response rate, safety, and tolerability.
The trial had statistical power to detect a 34% improvement in the hazard for overall survival (P=0.04) and a 35% improvement in the hazard for progression-free survival (P=0.01).
The study population had a median age of 53 to 54, and all had good performance status. O'Shaughnessy reported that 56% to 57% of patients were receiving first-line therapy for metastatic disease, and 43% to 44% were being treated in second or third line. Patients in the placebo group had a median disease-free interval of 15 months compared with 12 months for the iniparib group.
The rates and types of treatment-emergent adverse events were similar between the two groups, suggesting that iniparib did not add appreciably to the anticipated toxicity of the chemotherapy.
When the trial ended, the iniparib arm had a median progression-free survival of 5.1 months compared with 4.1 months in the placebo group. The associated probability value (P=0.027) did not meet the prespecified definition of P=0.01.
Median overall survival was 11.8 months with iniparib and 11.1 months in the placebo arm (P=0.28).
Analysis of secondary endpoints showed an objective response rate of 34% in the iniparib group and 30% in the placebo arm. The two groups had similar rates of complete and partial responses.
An exploratory analysis of the primary endpoints by treatment history showed that patients receiving first-line metastatic therapy had similar overall survival and progression-free survival.
In contrast, patients in second- and third-line therapy had a median progression-free survival of 4.2 months with iniparib and 2.5 months with placebo, a difference that translated into a hazard ratio of 0.57 in favor of iniparib.
Median overall survival was 10.8 months in the iniparib arm and 8.1 months in the placebo arm, representing a 35% reduction in the hazard.
A one-month improvement in progression-free survival with iniparib did not meet the prespecified definition of statistical significance. Adding the PARP inhibitor to chemotherapy was associated with less than a one-month improvement in overall survival compared with chemotherapy alone.
An exploratory analysis suggested that patients who received iniparib as second- or third-line therapy might have benefited from treatment, Joyce O'Shaughnessy, MD, reported at the American Society of Clinical Oncology meeting.
"Metastatic triple-negative breast cancer is a highly heterogeneous patient population on intrinsic subtyping," said O'Shaughnessy, of Baylor University Medical Center in Dallas. "Biomarker analyses are under way to evaluate patient populations that may benefit from iniparib."
PARP inhibitors block tumor cells' ability to repair DNA damage, including damage caused by chemotherapy. Preclinical studies have suggested the agents may also potentiate chemotherapy-induced DNA damage.
The drugs are being evaluated clinically in multiple tumor types. Earlier this year, O'Shaughnessy and colleagues reported final results of a phase II study showing significant improvement in progression-free survival in patients with metastatic triple-negative breast cancer treated with iniparib (N Engl J Med 2011; 364: 205-214).
However, less than a month after publication of the positive results sanofi-aventis announced that a larger phase III study had failed to corroborate the phase II results. O'Shaughnessy's presentation at ASCO confirmed the study's negative outcome.
The phase III trial involved 519 women with metastatic triple-negative breast cancer, treated with two or fewer metastatic regimens. All patients received gemcitabine and carboplatin and were randomized to iniparib or placebo. Patients who progressed on placebo could cross over to the iniparib arm, and 96% of patients in the placebo arm subsequently crossed over.
The trial had coprimary endpoints of overall survival and progression-free survival, and the study would be considered positive if either endpoint favored iniparib. Secondary endpoints included objective response rate, safety, and tolerability.
The trial had statistical power to detect a 34% improvement in the hazard for overall survival (P=0.04) and a 35% improvement in the hazard for progression-free survival (P=0.01).
The study population had a median age of 53 to 54, and all had good performance status. O'Shaughnessy reported that 56% to 57% of patients were receiving first-line therapy for metastatic disease, and 43% to 44% were being treated in second or third line. Patients in the placebo group had a median disease-free interval of 15 months compared with 12 months for the iniparib group.
The rates and types of treatment-emergent adverse events were similar between the two groups, suggesting that iniparib did not add appreciably to the anticipated toxicity of the chemotherapy.
When the trial ended, the iniparib arm had a median progression-free survival of 5.1 months compared with 4.1 months in the placebo group. The associated probability value (P=0.027) did not meet the prespecified definition of P=0.01.
Median overall survival was 11.8 months with iniparib and 11.1 months in the placebo arm (P=0.28).
Analysis of secondary endpoints showed an objective response rate of 34% in the iniparib group and 30% in the placebo arm. The two groups had similar rates of complete and partial responses.
An exploratory analysis of the primary endpoints by treatment history showed that patients receiving first-line metastatic therapy had similar overall survival and progression-free survival.
In contrast, patients in second- and third-line therapy had a median progression-free survival of 4.2 months with iniparib and 2.5 months with placebo, a difference that translated into a hazard ratio of 0.57 in favor of iniparib.
Median overall survival was 10.8 months in the iniparib arm and 8.1 months in the placebo arm, representing a 35% reduction in the hazard.
Δεν υπάρχουν σχόλια:
Δημοσίευση σχολίου