PARIS – The Worldwide Innovative Networking (WIN) Consortium at its annual symposium this week provided an update on its precision oncology clinical trial pipeline.
On the heels of reporting data from the WINTHER trial at the American Society of Clinical Oncology's annual meeting earlier this month, the consortium leadership this week discussed launching another study called MERCURY. Like WINTHER, this new trial will also use both DNA and RNA data to match patients with a variety of tumors types to targeted therapies and evaluate the impact of this approach on their outcomes.
The consortium has enrolled the first three patients into SPRING, a trial that is investigating the safety and efficacy of triplet therapies as a front-line option for advanced non-small cell lung cancer patients and will validate a biomarker algorithm for identifying which patients will benefit from these combinations. WIN is also close to launching a repository of blood and tissue samples, called BOOSTER, that will be used to discover and validate blood biomarkers for the early detection of NSCLC.
At ASCO, researchers presented data from WINTHER — WIN Consortium's first completed study — showing that the addition of transcriptomics to genomics allowed 35 percent of 303 consented advanced cancer patients to be matched to a treatment. Without the RNA-based algorithm, the match-rate would have been 23 percent.
Although the study did not meet its prespecified clinical benefit endpoint due to a number of factors, a blinded, post-hoc analysis showed that when patients received treatments they were most likely to benefit from as determined by high matching scores, they lived significantly longer compared to those who did not get the top-matched therapies. The study demonstrated the value of RNA data in precision oncology, the importance of gauging gene expression in tumor and normal samples, and that patients were willing to provide these samples.
WIN consortium is taking these lessons and incorporating them into MERCURY, a double-basket study that will enroll patients with a range of tumors, and match them to a variety of drug arms including single and combination agents based on genomic and transcriptomic profiling. Patients will be tested on a 270-gene next-generation sequencing panel and have their differential gene expression measured in normal and tumor tissues using HTG Molecular Diagnostics' nuclease assay.
The WINTHER algorithm will propose the best matches for each patient, and principal investigators will assign patients to a cohort guided by the match score and DNA data. For every drug/tumor arm, researchers will aim to enroll 10 patients in the first phase of MERCURY, and the arm will be expanded to the second phase only if two or more see their tumors shrink in the first four months of treatment.
Cohorts that expand to the second phase will enroll another 18 patients. Seven or more patients will have to have a response in order for a particular therapy to be considered effective, and the drugs that yield the best responses will move on for evaluation in Phase III registration trials.
WIN is hoping to demonstrate that the basket study design will result in more efficient development of cancer drugs by advancing treatments with a higher chance of success in Phase III. Highlighting MERCURY at the symposium, Richard Schilsky, chairman of WIN and chief medical officer of ASCO, noted that the attrition rate in oncology drug development is 82 percent and around 53 percent of kinase inhibitors fail. Moreover, in Phase III studies, 45 percent of drugs fail due to lack of efficacy and 24 percent don't succeed because they didn't extend survival compared to existing options.
The lessons from the WINTHER trial are also important for SPRING, another WIN study the US Food and Drug Administration approved for launch last year. The proof-of-concept study will evaluate the safety and efficacy of triple therapies in advanced NSCLC patients, and retrospectively validate the ability of the SIMS algorithm, which integrates genomics and transcriptomics, to match patients to a combination therapeutic strategy.
In a 2015 Oncotarget paper, researchers led by Razelle Kurzrock, director of the Center for Personalized Cancer Therapy & Clinical Trials Office at the University of California, San Diego's Moores Cancer Center, described how they developed SIMS by analyzing lung tumor and normal tissue using targeted genomic sequencing and also looking at copy number variation, transcriptomics, and miRNA expression.
Using these techniques, they identified 24 "interventional nodes" that were driving cancer but could also be targeted by drugs, and developed a scoring system to rank how disturbed these nodes were. Based on which nodes were co-activated, Kurzrock's group defined triplet therapies for overcoming resistance.
The transcriptomic algorithm used in the WINTHER study was a precursor to SIMS. "Today, personalized oncology is based mostly on DNA sequencing. We believe … genomics is the tip of the iceberg," Kurzrock said at the symposium this week. "The idea is to go a little deeper into that iceberg and incorporate transcriptomics using comparison of tumor and normal tissue."
She estimated that SPRING will need to study around 10 to 12 combinations to ensure matches for most NSCLC patients. The first triplet being investigated in SPRING include two treatments developed by Pfizer, Ibrance (palbociclib) and Inlyta (axitinib); and a third, Bavencio (avelumab) codeveloped by Pfizer and Merck KGaA. Ibrance is a CDK4 and CDK6 inhibitor. Inlyta is anti-angiogenic drug that blocks c-KIT, VEGF receptors, and PDGFR, while Bavencio is an immunotherapy that targets PD-L1.
The hypothesis behind SPRING is that "drugs with different modalities of action will be able to achieve what single agents can't achieve," said Kurzrock, adding that if patients don't respond to one triplet, they could rotate to another arm. Each arm will operate in two phases: patients will undergo dose escalation in Phase I and receive treatment at the recommended doses in Phase II.
In the Ibrance/Inlyta/Bavencio arm, the plan is to enroll 100 patients in the Phase II portion. If in the initial crop of 50 patients retrospective analysis of SIMS shows that the algorithm can accurately predict responders, the researchers may ask FDA for permission to prospectively use SIMS to match the rest of the cohort, or an expansion cohort, to the triplet.
In order to be eligible for SPRING, however, advanced NSCLC patients can't already have targetable genomic alterations in EGFR, ALK, or ROS1 genes, for example. Moreover, they will receive the triplet combination in the first-line setting.
"This trial was developed with input from the FDA," Kurzrock said. "They actually encouraged us to include patients in the first line … because they felt that treatments for patients that don't have [targetable] tumor drivers are still not good enough."
So far the Ibrance/inlyta/Bavencio arm has enrolled three patients. One patient has completed the dose escalation phase. Benjamin Solomon from Avera Cancer Institute and a SPRING investigator said that the other two patients enrolled at that institute are at the first dose level and are tolerating it well. "That was expected based on what we know about overlapping toxicities in these three drugs," Solomon said.
He added that as seen in WINTHER, these patients also seem willing to undergo dual biopsies once they understand the value of it. "Our two patients who have been enrolled haven't had any complications from that," Solomon said.
Meanwhile, as the first SPRING arm continues to enroll patients, the consortium already has a second arm waiting in the wings, which will investigate Inlyta, Bavencio, and Merck KGaA's c-Met inhibitor tepotinib.
Another area WIN is hoping to make an impact is in the area of blood-based diagnostic markers for NSCLC. With BOOSTER, "we want to look for biomarkers for early detection [of NSCLC] and monitor the course of the disease," said Amir Onn, a lead investigator in the effort who also heads up the Institute of Pulmonary Oncology at Sheba Medical Center in Israel.
Despite the positive impact of targeted drugs in certain molecularly defined subpopulations of cancer patients, the mortality rates in NSCLC have remained stubbornly low. Around 14 percent of NSCLC patients with stage IIIA disease are alive five years after diagnosis; with stage IIIB disease the five-year survival rate is 5 percent. When the disease has spread to other parts of the body, the five-year survival rate drops to 1 percent.
"There is a need to do better in diagnosing and managing this disease," Onn said.
Moreover, current tools are not good at diagnosing lung cancer. In the US, it takes 350 CT scans to detect one case of lung cancer, Onn noted. He also cited a study in which US Veterans Affairs hospitals implemented lung cancer screening guidelines, and screened some 2,000 patients who consented to partake in the study. CT scans detected nodules in 60 percent of patients, the vast majority of which were benign.
"So, obviously, CT is not good enough to detect lung cancer," Onn said. "We need additional technologies. Why don't we use blood or tissue markers?"
However, he cited another study that demonstrated the ability of a multi-analyte blood test to detect a range of cancers. It didn't work as well in lung cancer compared to its performance in other tumor types. The WIN Consortium is hoping to make a difference in this regard through BOOSTER.
This effort is planning to enroll 4,000 NSCLC patients globally who undergo curative surgery for early-stage lung cancer, and collect blood before and after three months following surgery. These samples will be compared for identification of a variety of biomarkers. "Hopefully, we'll be able to identify markers that are associated with disease," Onn said. The researchers then plan to validate these biomarkers by following patients longitudinally to see whether or not they develop the disease.