Optimizing Oncology Clinical Trials with Personalized Medicine
Personalized medicine is the art and science of “coupling established clinical–pathological indexes with state-of-the-art molecular profiling to create diagnostic, prognostic and therapeutic strategies precisely tailored to each patient’s requirements.”1 The clinical development and regulatory approval of targeted agents that have therapeutic benefit in molecularly defined patient subsets has highlighted the value of personalized medicine, which can be leveraged in clinical trials so that the identification of eligible patients for certain trials is optimized based on the patient’s molecular testing. These targeted patient groups are often rare and expensive to identify, particularly in the field of oncology, making clinical trials targeting these populations difficult to enroll.
Patient pre-profiling, determining a patient’s tumor-specific molecular profile, could have great value for physicians and patients as they consider potential clinical trial options. To the extent that a molecular profile is broad, encompassing many potential target alterations, one can easily imagine benefits such as more efficient enrollment, increased participation by physicians and patients and more informed treatment decisions. Pre-profiling before a trial is set up at a particular site offers a more efficient approach to enrollment while also providing greater molecular testing to support patient care.
A growing proportion of drugs in clinical development target molecular mechanisms and niche patient populations defined by the presence of the altered target biomarker in the patients’ tumor.2 As a result, molecular-based patient testing to identify patients for study enrollment is becoming more common place in centers performing clinical research. Increasingly, there is the need to identify small fractions of a clinico-pathologically defined patient subset. For example, there is great interest in identifying non-small-cell lung carcinoma (NSCLC) patients harboring a ROS1 translocation. However, these patients represent only 2–3% of the adenocarcinoma NSCLC population. Identifying these patients is becoming a significant challenge for drug developers as this often leads to increased costs and prolongs enrollment times for these clinical studies.
Current practices for identification and enrollment of molecularly-defined subsets of cancer patients are inefficient. A physician may perform a screening diagnostic test (for EGFR mutation status, for example) and if that does not permit enrollment into the trial of interest, they may decide on a standard of care or screen, using a separate diagnostic test (ALK mutation, for example) for a second clinical trial. In addition to the cost and time concerns already noted, the sequential screening process is certainly not in the patient’s best interest in some common clinical situations where oncologic disease can progress rapidly. It should also be noted that often a limited amount of tumor tissue is available for diagnostic testing, limiting the number of diagnostic tests that can practically be performed. The effect of all of these factors is that identification of these small subsets of patients is costly, takes too much time and reduces efficiency in clinical trial enrollment and participation.
The noteworthy recent codevelopment and commercial success of targeted agents with companion diagnostics, including Xalkori (crizotinib) and Zelboraf (vemurafenib), suggest that the era of personalized medicine has arrived and is benefiting patients. Molecular, and especially genomic, information is increasingly being used for patient diagnosis, estimation of prognosis and selection of targeted, safe and effective treatment of cancer patients. While a personalized medicine approach promises to provide safer and more efficacious patient care, application of these approaches may also have the potential to speed enrollment for trials targeting molecularly-defined patient populations.
It is not surprising that some academic-based oncology practices have begun to perform broad-based molecular diagnostic testing on most of the patients entering their practice3,4 to take advantage of emerging genomic technologies and the personalized medicine approach. This personalized medicine-inspired profiling is already yielding benefits for patients.4,5 The potential of this approach is made tangible by the development of broad-based genomic testing of tumors as well as recent examples of novel trial designs that leverage genomic profiles.
The availability of broad-based genomic testing of tumors that can routinely provide genomic information regarding a focused set of genomic alterations (25–50 genes) or a comprehensive set of alterations (200 to whole-exon sequencing) has made genomic testing relevant to contemporary cancer care in the context of the treatment and clinical trial enrollment decisions routinely confronting physicians. Such testing is available at some academic medical centers, large oncology networks and practices and through commercial companies.
This sets the stage for novel clinical trial designs that can better utilize such genomic information. For example, novel designs represented in both the current BATTLE-2 and ISPY-2 trials—in which multiple drugs in various arms of the study target different molecular alterations6,7—demonstrate the value of broad-based genomic profiling of patient tumors. These clinical studies successfully target specific clinic-pathologic sets of patients.
Others have noted, however, that a single genomic alteration or pathway might have clinical relevance across clinical indications. In this context, the concept of a “basket study” has been developed.8 In this clinical trial design, patients of most any clinical classification whose tumor contains the genomic alteration may be eligible. In addition, an approach termed a “master protocol”9 seeks to evaluate several targeted agents simultaneously following biomarker screening in the setting of squamous cell NSCLC, building on the success of the BATTLE trial.
All of these efforts seek the same primary goal: speed the enrollment of clinical trials and efficiently match patients to drugs, thus speeding the development of new targeted cancer agents. Improving the enrollment rate of patients in oncology trials is a goal shared by all stakeholders. Moreover, access to this information by physicians could allow for more informed treatment choices for their patients resulting in improved patient outcomes as evidenced by the studies cited previously.
There are several factors critical for the success of a preprofiling program as previously described. A primary, initial consideration is how the preprofiling will be funded and how the business model between the sponsor(s), contract research organization (CRO) and investigator sites will be structured. Various options may be considered, ranging from a combination of government and private funding sources, sponsor-funded screening or patient registries, to value-based pricing based upon study performance milestones. It is anticipated that several models will be explored and the ones that dominate will deliver real value to the various primary stakeholders (patients, investigators, sponsors and CROs). Other key success factors include mitigation of any regulatory, ethical or legal risks.
- A successful preprofiling program will enable the following:
- Improved access to, recruitment and retention of profiled and qualified patients matched to the appropriate clinical trials;
- Streamlined processes integrating clinical and genomic data for patient trial matching;
- Optimized patient sample management and use;
- Integration of molecular profiles and bioinformatic tools with sponsor access to a centralized data repository for greater data sharing and utilization;
- Potential for shared operational processes between organizations (for example, site and networks, biopharmaceutical sponsors, CROs and payers);
- Creation of patient registries for selected diseases or indications including clinically annotated samples with molecular profiles supporting analyses and outcomes research;
- Rapid turnaround of testing and reports.
Preprofiling and trial enrollment in oncology is best utilized in the context of an operational structure that has access to a large menu of trials, patients and investigator sites. Development of these sorts of relationships between a CRO/sponsor and sites will be an important component of preprofiling. There is evidence in the biopharma field that these sorts of relationships are already under development.10,11
Our current understanding of cancer biology indicates that cancer is a large number of niche diseases that may be targeted with therapies against specific molecular alterations. Drug development under this model creates new challenges for both the development program itself and for patient care. Patient preprofiling promises to leverage high throughput genomic profiling, bioinformatics and, where possible, new trial designs to drive better trial matching and faster enrollment to clinical trials.
Preprofiling may require new relationships and business models, most notably, among patients, clinical sites, biopharmaceutical sponsors and CROs to enable data sharing, site start-up and funding of the genomic testing. The adoption of a new model of patient genomic profiling linked to novel clinical trial designs testing targeted therapies in development is becoming a key response to the challenge to develop many compounds in niche populations in a cost and time sensitive manner.
1. Mirnezami R, Nicholson J, Darzi A. Preparing for precision medicine. N Engl J Med. 2012;366:489-491.
2. Shelley S. Biomarkers and companion diagnostics expand drug potential. Pharmaceutical Commerce. September 2013.
3. Roychowdhury S, Iyer MK, Robinson DR, et al. Personalized oncology through integrative high-throughput sequencing: a pilot study. Sci Transl Med. 2011;3:111ra121.
4. Von Hoff DD, Stephenson JJ Jr, Rosen P, et al. Pilot study using molecular profiling of patients’ tumors to find potential targets and select treatments for their refractory cancers. J Clin Oncol. 2010;28:4877-4883.
5. Tsimberidou AM, Iskander NG, Hong DS, et al. Personalized medicine in a phase I clinical trials program: The M. D. Anderson Cancer Center Initiative. J Clin Oncol. 2011; 29: (suppl; abstr CRA2500).
6. Kim ES, Herbst RS, Wistuba II, et al. The BATTLE trial: personalizing therapy for lung cancer. Cancer Discov. 2011;1:44-53.
7. Barker AD, Sigman CC, Kelloff GJ, et al. I-SPY 2: an adaptive breast cancer trial design in the setting of neoadjuvant chemotherapy. Clin Pharmacol Ther. 2009;86:97-100.
8. Willyard C. ‘Basket studies’ will hold intricate data for cancer drug approvals. Nat Med. 2013;19:655.
9. Ledford H. ‘Master protocol’ aims to revamp cancer trials. Nature. 2013;498(7453):146-147.
10. Sarah Cannon Research Institute. Sarah Cannon Research Institute and AstraZeneca announce personalized medicine partnership and enhanced strategic clinical development collaboration. Sarah
Cannon Research Institute. http://sarahcannon.com/about/newsroom/sarah-cannon-research-institute-and-astrazeneca-announce. Published July 1, 2013. Accessed November 13, 2013.
11. Quintiles Inc. Quintiles advances new approach to speed biomarker-targeted therapies to cancer patients. Quintiles Inc. http://investors.quintiles.com/investor-relations/press-releases/press-releases-details/2013/Quintiles-Advances-New-Approach-to-Speed-Biomarker-Targeted-Therapies-to-Cancer-Patients/default.aspx. Published September 24, 2013. Accessed November 13, 2013.