Friday morning opened with two trials using carboplatin in triple negative breast cancer in combination with one other drug. The trial comparison was difficult due to differences in patient populations and the use of additional drugs; in one case carboplatin was paired with veliparib and in the other bevacizumab (Avastin).
Dr. William Sikov (Miriam Hospital, Providence) presented for the CALGB/Alliance a 2×2 study design with four distinct arms: all patients received standard chemo paclitaxel and AC; half of these were randomized to receive carboplatin; and half of both arms were then randomized to receive bevacizumab resulting in four distinct treatment arms. The majority of patients were between 40-59 years old and a full 20% of trial participants were African American.
The study looked at pCR (pathological complete response) and found that 60% of patients who received carboplatin achieved pCR compared to 46% of patients who did not receive carboplatin. The patients who received bevacizumab also had higher pCR attainment, but not as high (59% compared to 48%) and not statistically significant pCR for both breast and axilla, which is the standard definition. The treatment arm that received both carboplatin and bevacizumab in combination achieved a pCR 67% but not statistically significant and the researchers couldn’t find a combination effect.
Toxicity was also worse with the additional drugs. Carboplatin increased the rates of Grade 3 side effects and led to more nausea and vomiting. Only patients with both carboplatin and bevacizumab experienced neutropenia (abnormally low white blood cells).
In light of the adverse effects, some life threatening, of bevacizumab any benefit in pCR is outweighed by its increased toxicity. So balancing benefit and risks—not to mention cost—it is not recommended. Indeed, although there have been several trials (though my no means all) which suggest improved response in metastatic breast cancer, they have not translated to overall survival. Data continues to show no survival benefit for Avastin and high toxicity. If a subset of patients benefit, we still don’t know which ones.
In contrast, carboplatin increased pCR in both breast and breast/axilla. While this is an early trial and optimal dose and schedule is not known, there may be movement toward using carboplatin as part of standard neoadjuvant therapy for stage II-III triple negative breast cancer along with a standard chemo backbone. Note that data from other trials has shown the benefit of carboplatin for cancers associated with a BRCA mutation.
The second trial had a very interesting and novel trial design. Dr. Hope Rugo, from University of California, San Francisco, presented data from I-SPY 2, which introduces several new agents and adjusts enrollment using an automated algorithm. Randomization is based on learning and adaptation as the study progresses. This allows the researchers to add and drop agents as they learn about effectiveness of agents in specific subtypes through the automated algorithm.
The goal of I-SPY 2 is to identify promising drugs and using pathological complete response (pCR) to accelerate the process. Today Dr. Rugo reported on one of the total seven experimental arms of a Phase II trial.
All patients in I-SPY 2 receive paclitaxel with either the experimental agent or control. If HER2+, patients receive pertuzumab. Randomization is based on biomarker subtypes/signatures: there are a total of 8 subtypes. The first treatment to “graduate” the trial is veliparib and carboplatin added to paclitaxel followed by AC, for treatment of triple negative breast cancer.
This small Phase II trial included just 72 patients randomized to get the valiparib and carboplatin experimental arm. These patients on the experimental arm achieved superior pCR at 52% of patients. While more patients in the experimental arm of the study discontinued treatment early because of side effects, in the control arm of the study more patients discontinued treatment because of progression of their cancer. Dr. Rugo suggested that hematolitic issues (destruction of red blood cells) on the experimental arm could be managed by reduction and delay of the dose. She therefore recommends veliparib with carboplatin for further research.
The result of three separate randomized neoadjuvant studies have now established that inclusion of carboplatin increases pCR in triple negative disease. However, survival has yet to be shown. Furthermore, it is not possible to identify which drug in the carboplatin combinations (or perhaps the combination itself) produces the gain. Furthermore dose and schedule of treatment still remain to be established.