New Clinical Trials Designed to Better Leverage Precision Medicine
By Jennifer Levin Carter, MD, MPH, Chief Medical Officer and Founder, N-of-One
Next Generation Sequencing is leading to the discovery and approval of a growing number of effective targeted cancer therapeutics that are based on uncovering the molecular characteristics of tumors. Although the number of potential targets continues to grow, drug approvals for targeted therapies have lagged behind these biologic discoveries, in large part due to the need for new clinical trial designs. Historically, clinical trials have been designed to treat cancers classified by their anatomic location, without consideration of molecular alterations that may be present in subsets of such tumors.
Oncology drug development is one of the largest therapeutic areas for pharmaceutical companies. To get a single oncology drug through clinical trials and the FDA approval process, pharmaceutical companies can spend over $1 billion, and the process can take12 to 15 years and require thousands of patient volunteers (Foundation for National Institutes of Health). In addition, clinical trial success rates have been very low: 70% of oncology drugs fail in Phase II and 59% of oncology drugs fail in Phase 3 (Kola, 2004).
A new paradigm is clearly needed. It has been hypothesized and demonstrated that using molecular markers to stratify patients into appropriate clinical trials can improve success rates of trials. While a study analyzing clinical trial drug development in breast cancer from 1998 to 2012 found that only 14% of drugs in clinical development for advanced disease were approved, when patients were selected based on HER2 status, the success rate increased to 23%. Using HER2 status to stratify patients also led to a 27% decrease in costs (Parker, 2012).
Similarly, clinical trial success rates in NSCLC during the same time period were six times higher for biomarker-targeted therapies and three times higher for receptor-targeted therapies than for trials that were not molecularly guided. The risk-adjusted cost for NSCLC clinical drug development was estimated to be nearly $2 billion (Parker, 2014).
Over the past few years, two new trial methodologies have emerged in order to address this growing need for a new clinical trial approach:
- The adaptive design and
- The biomarker-guided design.
The goals of these two methods are three-fold:
- More effectively match drugs to the populations most likely to benefit based on an individual patient’s molecular profile
- Decrease trial costs
- Decrease the time to drug approval.
The strategies involved within these new designs include interim trial analyses of individual patient data, as well as pre-trial biomarker screening combined with a deeper analysis of the molecular variability within individual patient tumors. The FDA has drafted industry guidance outlining the use and importance of adaptive trials, and various biomarker status initiatives are underway or are being developed.
The Adaptive Design Clinical Trial
The purpose of the adaptive trial design is both to increase the likelihood of a study’s success as well as to deliver more in-depth data concerning the treatment’s effects. The FDA defines an adaptive trial as “a study that includes a prospectively planned opportunity for modification of one or more specified aspects of the study design and hypotheses based on analysis of data (usually interim data) from subjects in the study.”
A key feature of this trial design is the use of interim analyses, which permit researchers to customize the trial, for example, by adjusting the randomization process or treatment dose or schedule, based on results from earlier participants in the study. This allows for potentially ineffective parameters within the trial to be minimized, instead of being locked into the study’s original design (FDA, 2010).
There are more than 20 ongoing adaptive trials in the pipeline. Two examples that have yielded recent positive results are the I-SPY 2 and BATTLE trials.
The I-SPY 2 trial, jointly conducted by the NCI, FDA and Biomarkers Consortium for patients with newly diagnosed, locally advanced breast cancer, was launched in 2010. The trial screens up to 12 cancer drugs from multiple pharmaceutical companies over the duration of the trial, by adding each individually to a standard neoadjuvant chemotherapy backbone. The adaptive approach in this study involves performing transcriptional profiling on patients’ tumor samples upon trial entry, and calculating incoming patients’ probable responses to the trial’s various therapies based on how previous patients’ tumors with similar genetic signatures responded to these treatments. The outcomes are then used to weight the randomization that assigns each patient to a trial arm (Printz, 2013).
Attractive features of this trial’s design include the relatively rapid assessment of the primary endpoint (approximately five months after initiation of neoadjuvant treatment), the evaluation of multiple therapeutic agents with one control arm, and the ability to predict benefit in molecularly defined subsets, allowing the design of future larger trials within the molecularly defined group of patients (Berry et al, 2012 – Nat Rev Clin Oncol).
Recently, two of the drugs involved, veliparib (AbbVie Inc.) and neratinib (Puma Biotechnology Inc.), were found to have promising results in two different breast cancer subtypes. The veliparib-containing regimen was estimated to have a pathologic complete remission (pCR) rate of 52% in patients with triple negative disease, compared to 26% for patients treated with standard chemotherapy alone (SABCS, 2013).
For the neratinib-containing regimen, it was predicted that if tested in a Phase 3 trial, this regimen has a 94.7% Bayesian probability of proving superior to standard therapy in women with HER2 positive/hormone receptor negative disease (Puma Biotech, 04/12/2013). On January 15, 2014, AbbVie initiated a Phase 3 trial for veliparib, and Puma Biotechnology is currently discussing plans for a Phase 3 trial for neratinib.
The MD Anderson Cancer Center’s Biomarker-integrated Approaches of Targeted Therapy for Lung Cancer Elimination (BATTLE) trial is another adaptive trial that recently reported positive data. However in this trial, multiple drugs are tested on multiple mutations within a single cancer type. In the first phase, ~40% of patients had biomarker testing and were randomly assigned to one of the 4 treatment arms. The second phase of the trial was adaptive, in which the remaining 60% were assigned to treatments based on their biomarker status and how patients with similar tumor profiles had responded in the first phase of the trial.
In December 2013, the BATTLE investigators reported that sorafenib demonstrated clinical activity in NSCLC patients, particularly in those harboring wild-type EGFR: the eight-week disease control rate (DCR) was 58.2% (98 patients total), and the median progression-free survival (PFS) and overall survival (OS) were 2.83 and 8.48 months, respectively (Blumenschein, 2013).
By contrast, in May 2013 the vandetanib arm of the trial was reported to have results that were similar to those reported in the literature for unselected NSCLC populations (Tsao, 2013). The remaining two trials are currently in the pipeline.
New Clinical Trials Designed to Better Leverage Precision Medicine (cont.)
The Biomarker Status Trial
Biomarker status trials’ unique aspect is that they use patient biomarker status across multiple tumor types to randomize patients to treatment arms of a trial.
Basket/Bucket Trials: NCI-MATCH
The basket or bucket-trial method recruits patients via biomarker status instead of cancer type. After biomarker identification, the patients are divided into multiple study arms (baskets) by cancer type, and the drug’s impact is assessed within the separate arms as well as within the study as a whole (AACR, 2013).
Researchers hope that basket studies will also help accelerate the drug approval process for treatments targeting rare cancer types for which large randomized trials are logistically impossible. If a drug were shown to be active in in one of these types of cancer, the basket trial would allow for the enrichment of that particular cohort (Willyard, 2013).
Various studies are underway to test the efficacy of this type of approach. Two examples are Roche’s Phase 2 study of vemurafenib in patients with BRAF V600-mutant cancers (NCT01524978) and the Italian Sarcoma Group’s Phase 2 study of imatinib in patients whose tumors express the platelet-derived growth factor receptor (PDGFR) (NCT00928525). An additional study now within the planning stage is NCI-MATCH, which proposes to match 1,000 patients to various therapies based on biomarker status. So far, over 40 drugs have been pledged from companies such as Clovis, Genentech and Pfizer, and 20 separate arms have been established for tumors with genetic alterations in known tumor drivers such as EGFR, HER2, BRAF, and ALK.
To date, oncology drugs are approved for an indication based on cancer type, but MATCH hopes to deliver positive results that could support future approvals based on molecular targets as well.
Novartis SIGNATURE Trial
Another example of this approach is Novartis’ Phase 2 non-randomized, open-label SIGNATURE trial, which will assign patients with solid tumors to one of five different trial arms with different targeted therapy regimens. Patients will first have their tumors tested for the pertinent genetic alterations or biomarkers of pathway activation at a CLIA certified laboratory, and then will be assigned to a trial arm via the recommendation of a board of experts who have reviewed the data.
The design will provide for “matching of the genetic abnormalities found in their tumors to relevant drugs; rapid enrollment; little if any geographical constraints on participation; and, of course, access to promising new therapies that have already undergone safety testing,” Dr. Steven Stein, SVP of U.S. Clinical Development and Medical Affairs at Novartis, said in a statement to Cancer Commons.
Important to the success of these novel trial designs is the need for cooperation and collaboration between different pharmaceutical companies. Companies have begun to acknowledge this requirement as demonstrated by the recent collaboration of global CRO Aptiv Solutions with Novartis, Janssen Pharmaceuticals Inc., and Eli Lilly.
On February 19, 2014, these companies partnered to develop the ADDPLAN® DF Consortium, the purpose of which is to improve upon current standard of practice procedures for designing and analyzing Phase 2 dose-selection trials. The consortium identifies poor dose selection as one of the leading causes of Phase 3 trial failures, and plans to develop new adaptive statistical methodologies in order to reduce selection error and uncertainty (Aptiv Solutions, 02/19/2014).
Additional new strategies are being launched, such as The Master Protocol initially for squamous cell lung cancer. The trial is unique as it is a collaboration between lung cancer researchers, the National Cancer Institute, the Foundation of the National Institutes of Health, the Food and Drug Administration, Friends of Cancer Research, and industry (ASCO Post, 2013).
The Master Protocol investigates multiple new therapies simultaneously for one tumor type. Unlike I-SPY 2 and BATTLE in which markers and drugs are analyzed for association, each biomarker included is associated with a targeted therapy and assignment to that treatment is based on results of a validated diagnostic test. It is designed as a Phase 2/3 trial powered to demonstrate efficacy, with the goal of expediting FDA approval (FOCR, 2013).
Designing the Future
Through advances in Next Generation Sequencing technologies, it is becoming increasingly feasible to quickly and efficiently generate a molecular profile of each patient’s tumor. With this information, patients can now select to participate in clinical trials that test therapeutic response not only in the context of their specific disease but also in the context of their tumor’s molecular characteristics. Under this paradigm, data from clinical trials can be used more expeditiously to adjust participation and improve clinical trial metrics.
About the Contributor
Chief Medical Officer and Founder, N-of-One, Jennifer Carter, MD has been an early driver in shaping and delivering personalized medicine for oncologists at the point of care. Today, N-of-One is the leading provider of molecular interpretation and therapeutic strategies for oncology.
- AACR. Molecularly informed clinical trials. AACR Cancer Progress Report. 2013. <http://cancerprogressreport.org/2013/Pages/mol_clin_trials.aspx>
- AbbVie. AbbVie announces initiation of pivotal phase 3 study of veliparib (ABT-888) for patients with early-stage triple-negative breast cancer. <http://abbvie.mediaroom.com/2014-01-15-AbbVie-Announces-Initiation-of-Pivotal-Phase-3-Study-of-Veliparib-ABT-888-for-Patients-with-Early-Stage-Triple-Negative-Breast-Cancer>
- Aptiv Solutions. Novartis, Janssen Pharmaceuticals, Eli Lilly and Aptiv Solutions form a consortium to develop technologies for design and execution of ddaptive dose finding trials. <http://www.aptivsolutions.com/press-release/2014/02/novartis-janssen-pharmaceuticals-eli-lilly-and-aptiv-solutions-form-a-consortium-to-develop-technologies-for-design-and-execution-of-adaptive-dose-finding-trials/>
- Bianchi, Adam. Oncology clinical trials drug development resources and case studies. 2013 <www.cuttingedgeinfo.com>
- Blumenschein G, Saintigny P, Liu S, Kim E, Tsao A, Herbst R, Alden C, Lee J, et al. Comprehensive biomarker analysis and final efficacy results of sorafenib in the BATTLE Trial. Clinical Cancer Research. 2013 (24): 6967-75.
- Clinicaltrials.gov; NCT00928525
- Clinicaltrials.gov; NCT01495247
- Clinicaltrials.gov; NCT01524978
- Clinicaltrials.gov; NCT01711398
- EMA. Draft qualification opinion of MCP-Mod as an efficient statistical methodology for model-based design and analysis of phase II dose finding studies under model uncertainty. <http://www.ema.europa.eu/docs/en_GB/document_library/Regulatory_and_procedural_guideline/2013/10/WC500152067.pdf>
- Falconi A, Lopes G, Parker J. Biomarkers and receptor targeted therapies reduce clinical trial risk in non-small-cell lung cancer. J Thoracic Oncol. 2014(2):163-9.
- FDA Guidance for Industry. Adaptive design clinical trials for drugs and biologics. <http://www.fda.gov/downloads/Drugs/…/Guidances/ucm201790.pdf>
- Helwick C. Master protocol could revolutionalize trials in lung cancer, and eventually other cancers. ASCO Post. 11/01/2013. <http://www.ascopost.com/issues/november-1,-2013/master-protocol-could-revolutionalize-trials-in-lung-cancer,-and-eventually-other-cancers.aspx>
- Herbst R. Design of a disease-specific master protocol. Conference on Clinical Cancer Research. 11/2012. <http://www.focr.org/sites/default/files/CCCR12MasterProtocol.pdf>
- NIH. Foundation for the NIH announces first results of I-SPY 2 breast cancer clinical trial. Foundation for the NIH. 12/26/2013. <http://www.fnih.org/press/releases/foundation-nih-announces-first-results-i-spy-2-breast-cancer-clinical-trial>
- Kim E, Herbst R, Wistuba I, Lee J, Blumenschein G, Tsao A, Stewart D, Hicks M, et al. The BATTLE trial: Personalizing therapy for lung cancer. Cancer Discovery, 2011;(1):44-53.
- Kola I, Landis J. Can the pharmaceutical industry reduce attrition rates? Nat Rev Drug Discov. 2004 Aug;3(8):711-5.
- Parker J, Lushina N, Bal P, Petrella T, Dent R, Lopes G. Impact of biomarkers on clinical trial risk in breast cancer. Breast Cancer Res Treat. 2012; Nov;136(1):179-85.
- Printz, C. I-SPY 2 may change how clinical trials are conducted. Cancer. 2013; 119 (11):1925–27.
- Puma Biotechnology. Puma Biotechnology Reports Positive Top Line Data from I-SPY 2 TRIAL Neratinib Graduates from I-SPY 2 TRIAL. 04/12/2013. <http://www.pumabiotechnology.com/pr20131204.html>
SABCS. New presurgery combination therapy may improve outcomes for women with triple-negative breast cancer. 12/13/2013.
- Tsao A, Liu S, Lee J, Alden C, Blumenschein G, Herbst R, Davis S, Kim E, et al. Clinical and biomarker outcomes of the phase II vandetanib study from the BATTLE trial. J Thorac Oncol. 2013 May;8(5):658-61.
- Willyard C. Basket studies will hold intricate data for cancer drug approvals. Nature Medicine. 2013(6):655.