Placeholder Banner

Improving Oncology Clinical Trials

February 8, 2016
The 18th Annual BIO CEO & Investor Conference kicked off with a panel focused on oncology therapeutics. A group of clinical trial experts gathered Monday morning in New York to discuss how to improve success in drug development. In particular, they explored alternative models for clinical trial design to reach higher success rates.

As a lead-in to the challenge faced, preliminary data from BIO and BioMedTracker was presented on clinical development success rates [here, and upcoming data here]. The key takeaway: the most recent decade of data shows oncology R&D with the lowest probability of success out of 14 disease areas.

The moderator for the session, Ellen Sigal, PhD, Chairperson and Founder, Friends of Cancer Research, started off by getting right to heart of the issue: biomarkers will help clinical development success in oncology, but we also need a better system for qualifying and validating biomarkers.

One of the most complex, alternative clinical trial designs that incorporates biomarkers is the public-private collaboration called “Lung-MAP,” a program in which Dr. Sigal is involved.  Lung-MAP uses a multi-drug, targeted screening approach to match patients with sub-studies testing investigational new treatments based on their unique tumor profiles. [See the trial design below from]

Trial DesignCombining the efforts of multiple biopharma companies, NIH, and other centers, this trial design offers increased information and infrastructure sharing, better access for patients, and less time and money needed before investigational drugs can be tested.

On the subject of how useful biomarkers can be, Eric Rowinsky, MD, PhD, Executive Chairman of Rgenix, added that it is not straightforward. Some of the most important drugs might not have biomarkers, so it raises the question how can you get funding if it is seen as high risk without the biomarkers. Recent PD-1 antibodies have had great success without biomarkers.

Addressing the low overall probability of success, Joseph Sparano, MD, Albert Einstein College of Medicine, made sure to acknowledge that in spite of low probabilities of success, there is a positive story in the field:  cancer mortality rates are dropping substantially because of novel medicines on the market today.

Dr. Sparano is involved in another adaptive trial called NCI-Molecular Analysis for Therapy Choice (NCI-MATCH).  Started in 2015, NCI-MATCH is a 10-arm trial that matches patient tumor DNA with drugs that may target certain mutations in that DNA. Investigators plan to obtain tumor biopsy specimens from as many as 3,000 patients, across many sub-types, and anticipate that more than 20 treatments will ultimately be tested, each in a different arm of the trial.

The primary endpoint for the trial is objective response rate (ORR), or the percentage of tumors that shrink).

The topic of best endpoints to use in cancer was also discussed. Paul Biondi, SVP, Head of Business Development, Bristol-Myers Squibb, points out that we should have more than overall survival (OS) or progression free survival (PFS) as endpoints. Potentially, surrogate endpoints and patient-reported data could be utilized more effectively.

Does the regulatory agency understanding these needs? “Yes, indeed,” says Mr. Biondi, adding that discussions are ongoing. Andreas Dreps, PhD, Senior Vice President, Drug Development Services, ICON Consulting Services, explained that the FDA is also open minded to single arm trials.

Do academics truly understand the regulatory process and what patients want? Dr. Sparano asserted that the academic centers are highly scientific. The field is becoming more segmented and may require some level of incrementalism in order to address specific issues. This takes time, as incremental, exploratory trials introduce additional steps prior to the initiation of larger trials with regulatory requirements. Furthermore, these exploratory trials may not align with patient group expectations or desires.

Regarding current thinking on patient reported outcomes (PROs), the panelists indicated that they would like to see more PROs used in trials and in regulatory filings, but it is not easy to make a notable difference  in the current regulatory environment. Validating these PROs would be a benefit as the new drugs would be further differentiated from old drugs

In conclusion, the panel was asked what the biggest barriers are for oncology drug development and what might the President’s new Moonshot project focus on? Below are a few of the responses:

  • Strong consortiums and strong project management and accountability. There is a big need for collaboration due to the number of combinations that need to be explored. It is mathematically impossible to explore all combinations in all sub indications, even for a big pharma company.

  • On the scientific front, the major challenge is to analyze all available data. There is so much underutilized data because we do not know how to apply it to the clinical setting.

  • Funding is still an issue. Even small SBIR-level funding can help creative, higher-risk research, the type of research that leads to novel breakthroughs.

  • We also do not have “short-term” biomarkers. In cardiovascular disease, we can check blood pressure, but in cancer, there is no way to quickly see if something is working.