SAN DIEGO—A clinical trial using the concept of adaptive randomization to identify patients most likely to benefit from the pan-erbB/HER inhibitor neratinib pointed to hormone-receptor (HR)-negative/HER-positive status as a biomarker signature for the further development of the drug.
In the data reported here at the American Association for Cancer Research Annual Meeting (Abstract CT227), patients with HR-negative/HER-positive high-risk stage II/III breast cancer treated with neoadjuvant neratinib plus standard chemotherapy had a pathologic complete response (pCR) rate of 56 percent, versus 33 percent for patients treated with standard neoadjuvant chemotherapy alone.
And an HR-positive/HER-positive arm in the same trial also had a higher pCR rate with neratinib—30 vs. 17 percent.
However, a triple-negative cohort did not have a statistically significant improvement in pCR, at 38 percent for the neratinib-containing regimen vs. 31 percent for the control arm. But patients with MammoPrint-positive (ultra-high) disease might also benefit from the trial regimen, with pCR rates of 48 percent for neratinib vs. 29 percent for standard chemotherapy, the researchers said.
These data were from one arm of the I-SPY 2 trial (Investigation of Serial Studies to Predict Your Therapeutic Response with Imaging and Molecular Analysis 2), a multicenter, Phase II, neoadjuvant trial for high-risk stage II/III breast cancer. The trial is using adaptive randomization to evaluate seven novel agents added to standard chemotherapy across a matrix of biomarker subtypes based on HR, HER2, MammoPrint, and MammoPrint High Risk.
In such a trial, the probability of treatment assignment is adjusted according to assigned treatments of patients already in the trial.
John W. Park, MD, Professor of Medicine at the University of California, San Francisco, Helen Diller Family Comprehensive Cancer Center, who presented the results, said pCR, the trial's endpoint, was defined as no residual invasive cancer in the breast or lymph nodes.
Novel agents or regimens “graduate” in the trial if there is an 85 percent or greater probability that they will be superior to a standard regimen alone for at least one signature in a Phase III neoadjuvant trial.
Once a drug has graduated, accrual to that arm ceases although data continue to be analyzed, including toxicity data, Park said.
He added that final probability data are recalculated when patients have gone to surgery, so the final probabilities are not necessarily the same as what were presented at the AACR meeting.
If a regimen does not graduate, the trial concludes. But lessons are learned from the responses obtained, Park said, and future randomization is revised to reflect results thus far.
“This accelerates the process of identifying drugs that are effective for specific breast cancer subtypes—It reduces the cost, time, and number of patients needed to get effective drugs to market.”
The adaptive statistical design used in I-SPY 2 was developed by the trial's senior investigators, principal investigator Laura J. Esserman, MD, Professor of Surgery and Radiology and Director of the Carol Frank Buck Breast Care Center at UCSF; and Donald A. Berry, PhD, Professor of Biostatistics at the University of Texas MD Anderson Cancer Center.
“Clinical trial designs have to change to keep pace with the amazing advances being made in biology,” Esserman said. “I-SPY 2 may not be the final answer, but so far it has been successful in expanding the boundaries of clinical trials, making them more patient-friendly while preserving their scientific integrity.”
Neratinib is one of seven concurrent agents being tested in I-SPY 2, she explained. Veliparib has already completed the process, and trials with AMG 386, AMG 479, MK 2206, pertuzumab, and pertuzumab-TDM1 are ongoing. The goal is to match investigational regimens with patient subsets on the basis of biomarker signatures that benefit from the regimen.
The first I-SPY 2-tested regimen to graduate was carboplatin combined with veliparib, a poly-ADP-ribose (PARP) inhibitor. In a report at the 2013 San Antonio Breast Cancer Symposium, the combination had an estimated 52 percent pCR in breast cancer patients with the triple-negative biomarker signature, versus 26 percent for carboplatin alone.
In the neratinib portion of I-SPY 2, the algorithm randomly assigned 115 evaluable patients to receive the neratinib-containing regimen, while another 78 patients were randomly assigned to a control group receiving a regimen of standard chemotherapy. The controls' outcomes were shared among the various I-SPY 2 investigational agents.
All patients first received paclitaxel once a week for 12 weeks and doxorubicin and cyclophosphamide every two to three weeks for four cycles, followed by docetaxel, doxorubicin, and cyclophosphamide.
HER2-negative patients then received neratinib plus docetaxel followed by doxorubicin, and cyclophosphamide, or docetaxel alone followed by doxorubicin and cyclophosphamide.
HER2-positive patients received neratinib plus docetaxel followed by doxorubicin and cyclophosphamide, or the benchmark of trastuzumab plus docetaxel followed by doxorubicin and cyclophosphamide.
Neratinib graduated specifically in the HER2-positive/HR-negative subsets, Park reported, with an estimated pCR rate of 56 vs. 33 percent for the concurrent trastuzumab-containing control treatment.
Adverse Events Considered Manageable
Park said there were few significant adverse events. The most severe was diarrhea: 39 percent grade 3 and above for neratinib vs. four percent for the control group. Neutropenia was also higher with neratinib: 17 percent grade 3 and above for neratinib vs. 10 percent for controls.
Prophylactic loperamide was begun later in the trial, and Park said aggressive supportive care for diarrhea was recommended for future trials of neratinib.
Early discontinuation was higher for neratinib—34 vs. 18 percent. For controls, this was due mainly to disease progression (18 percent of all 78 patients) while for neratinib it was due mainly to toxicity (20 percent of all 116).
Importantly there was no evidence of delayed surgery, at 24 weeks from treatment to surgery for both the neratinib arm and controls, Parks reported.
Now being planned, he said, is I-SPY 3, a Phase III registration trial.
I-SPY 2 is sponsored by the non-profit QuantumLeap Healthcare Collaborative, funded by philanthropic groups and pharmaceutical companies.
The Discussant for the study, Lajos Pusztai MD, DPhil, Professor of Medicine, Chief of Breast Medical Oncology, and Co-Director of the Cancer Genetics Research Program at Yale Cancer Center, noted that multiple clinical trials have shown in HER2-positive patients that a regimen that improves pCR rates invariably increases survival as well when tested in the adjuvant setting.
“These data led the Food and Drug Administration to consider pathological complete response as an endpoint to accelerated approval for high-risk prostate cancer. And indeed, in 2013, the first drug was approved under this program indicated for HER2-positive disease, veliparib.”
(Pusztai was also the Discussant for the veliparib presentation at the San Antonio meeting.)
He congratulated the researchers who designed the I-SPY trial: “Its beauty, elegance, and originality lies in the fact that in the same trial you can prospectively test a drug in multiple subsets in parallel—we have never done this before.”
It is very reassuring to see that neratinib graduated with the ER-negative/HER2-positive subset, he said, but it is also somewhat sobering to see that despite promising preclinical data, this drug, like any other drugs that target the EGFR and HER2 pathway, did not really have an impact on the triple-negative group of patients.
“The really exciting part of I-SPY is yet to come,” Pusztai said, as results for the other five experimental drugs in the I-SPY 2 program are revealed.
Pusztai said that for the pharmaceutical industry, “I-SPY and similar neoadjuvant trials can rapidly identify ‘add on’ drugs that warrant larger confirmatory trials.”
For laboratory scientists, adaptive trials will aid in developing response markers with clinically relevant discriminating value—“a huge unmet need and formidable challenge.”
And to clinical investigators, he said “there is more than one equally effective way to achieve the same desired clinical outcome—such as pCR—in HER2-positive breast cancer. The challenge is defining the most cost-effective and patient-friendly combination treatment strategies and sequencing of therapies.”
In a video interview on the iPad edition of this issue with OT reporter Dan Keller, Dr. Park elaborates on the clinical implications of the results, and explains how adaptive trial design allows faster testing of drugs and helps define the tumor signatures (subtypes by biomarkers) most sensitive to specific drugs and regimens.
If you are not yet receiving our iPad issues, download the free Oncology Times app from the App Store today! Visit http://bit.ly/OT-iPadApp, search in the App Store, or follow the link on oncology-times.com.