Patient Shortage for Cancer Trials Grows

Anthony W. Tolcher, MD

At a time of rapid advancements in cancer biology and innovation, researchers are facing mounting challenges in finding enough patients to recruit for clinical trials. Although this has been difficult for many years, current trends in oncology drug development may be making the problem more acute.

The proliferation of targeted investigational drugs has both narrowed the pool of eligible patients and increased demand for them. In addition, trials are conducted more often at centrally located academic centers than at the community oncology offices where the vast majority of patients with cancer go for treatment. These factors make the study and approval of new drugs more expensive and time-consuming and leave many people who have treatment-resistant cancers without access to potentially life-saving therapies.

Table. Key Considerations for Expanding Clinical Trial Eligibility⁶

These challenges have spurred efforts to expand and diversify the pool of clinical trial participants. Medical organizations have called for the relaxation of eligibility criteria and investigators are working through community practice networks to reach more patients (Table). Genomic testing companies, analytics firms, and drug makers are developing programs that collect patients’ data and connect them to appropriate studies, while working to bring trials to patients wherever they are rather than requiring them to travel long distances to benefit from new treatments.

“If we look at the data, historically less than 3% to 5% of patients actually get enrolled in clinical trials, and that doesn’t even account for the lack of diversity of patients,” said Arturo Loaiza-Bonilla, MD, chief of medical oncology and medical director of research at Cancer Treatment Centers of America in Philadelphia. “There are issues with patients older than 65, with racial minorities, and with populations of patients who don’t have access to academic centers to get treatment. It’s a very important issue and something that I look forward to seeing improve over time.”

Studies have consistently found that fewer than 1 in 20 adults with cancer enroll in clinical trials.¹ And about one-fifth of drug studies accrue less than half of their target number of patients after 3 years, according to an analysis of phase II and phase III adult trials sponsored by the National Clinical Trials Network, a program of the National Cancer Institute (NCI).²

Competing for Patients

When a trial closes due to low accrual, the researchers cannot report on the results and advance the drug toward approval. Trials with greater risk of low accrual included those that required a biopsy and phase III randomized trials, as well as those that faced more competition from other trials for patients with the same condition.A fundamental problem is that many potentially eligible patients never even reach the point of learning about a trial and deciding whether to participate. About 90% of patients are treated in the community setting and only 10% at academic centers, yet trials tend to be sited at the centers, said Anthony W. Tolcher, MD, of Texas Oncology and CEO and director of clinical research at NEXT Oncology, both in San Antonio.

“There’s a disconnect between where the studies are and where the patients are,” he said. Previously, “the expectation was that patients would come from all the surrounding areas to go onto NCI studies. That was fine when the National Cancer Institute used to be in the business of drug discovery and first-in-human type clinical studies, as well as cooperative groups, but that’s all changed. Almost every single new drug comes from biotech and the pharmaceutical industry.”

The shift to industry-sponsored research has generated a steady stream of more effective agents, revolutionizing cancer care, but the abundance of trials has overtaxed the pool of eligible patients at cancer centers, Tolcher said. The National Institutes of Health’s ClinicalTrials. gov database lists more than 8600 active cancer drug trials in the United States, including over 1000 for PD-1/PD-L1 inhibitors.³ Also, more than 240 immuno-oncology drugs were in development in 2017, according to Pharmaceutical Research and Manufacturers of America.⁴

Some companies are developing variations of existing checkpoint inhibitors to grab market share or to use in combination therapies without relying on competitors’ products. “Every company wants one,” said Richard L. Schilsky, MD, chief medical officer of the American Society of Clinical Oncology (ASCO). The trend of “me-too” drug development has led to concerns that research time and trial patients are being diverted for business reasons with little benefit to the patient community.

“Over time, you’re not driving innovation, which is supposedly what we’re looking for—not just profit or generation of a new drug that is going to be in the same space,” Loaiza-Bonilla said. Historical experience suggests that having multiple similar drugs will not even lead to lower prices, Schilsky said.

Targeted Therapy Challenges

However, others argue that drug development should not be discouraged even if it’s potentially duplicative. “You want to create as many opportunities as possible. I don’t think anyone has been harmed by the fact that there are 4 or 5 PD-1 [pathway] inhibitors out there and there are some going through development,” Tolcher said. He cited the example of Merck’s pembrolizumab (Keytruda), saying that although multiple agents targeting the pathway have reached the clinic, it ended up overtaking its competitors. “We’ve got to be very careful about the idea of turning down me-too drugs, because you never know which way it’s all going to work. Let the market decide,” he said.Even as the flourishing of industry-led drug development stokes demand for clinical trial patients, improved survival rates have narrowed the pool of people interested in volunteering.

“Well in excess of 50% of newly diagnosed cancer patients are cured with currently available therapies like surgery, radiation, and conventional chemotherapy,” Schilsky said. “It’s actually even difficult to study new approaches in some of these patient populations because the results are so outstanding with standard care approaches that it’s hard to demonstrate a substantially greater improvement with a new drug. That said, that still leaves a lot of patients who need new therapies when the existing therapies are ultimately ineffective. That’s where the clinical trial enrollment is so important but also so disappointing.”

One result is that research focuses increasingly on treatment-resistant cancers with rare biomarkers. For example, Loxo Oncology’s highly anticipated tumor-agnostic drug larotrectinib targets TRK-fusion proteins that are found in less than 1% of patients. “To find one patient you probably have to screen 500, and you have to do next-generation sequencing in all of them. So that makes it very, very difficult to find the right number of patients for that study,” Loaiza-Bonilla said.

Researchers evaluated larotrectinib based on results in 55 patients from 3 studies.⁵ “It took them years to find just those 55 patients, and they had protocols open at locations all around the world,” Schilsky said. “It can be done, but it’s not easy and it’s certainly not inexpensive.”

Physicians who run clinical trials must overcome a number of other hurdles as well. Sponsors often set stringent eligibility criteria, excluding patients who have had previous immunotherapy treatment or who have brain metastases or other concurrent malignancies. Tolcher said he has been offered highly restricted trials that appeared unlikely to attract enough eligible patients and declined to run them at his clinics.

Biopsy requirements are also creating challenges more often, said Hossein Borghaei, DO, MS, chief of thoracic medical oncology at Fox Chase Cancer Center in Philadelphia. During one study he was running, the sponsor came back and asked for additional tissue samples from patients before they could start on the drug.

“I’m not going to go back to the patient and say I want you to go through yet another course of significant biopsy just because I want to get you on a study,” Borghaei said. “I either start screening the patient for another study or I just go to standard of care by that time, so you lose that patient for a clinical trial. I don’t think 10 or 15 years ago it was as big of an issue.”

Given the rapid pace of drug development, issues such as slow accrual, logistical problems, and poor initial results can doom a trial, contributing to the high noncompletion rates that have been observed. “At least in the world of lung cancer, things are moving so rapidly that if you don’t complete your study in 12 to 18 months, the results might be irrelevant, because 3 other studies are reported in that period that completely change what we do as the standard of care,” Borghaei said. He said one study he joined was delayed between phase I and phase II while he figured out dosing schedules, but the company stopped making the drug in the interim, forcing the trial to close.

At the same time, investigators pointed to the improvements in survival rates enabled by clinical trials in recent years and said they did not believe enrollment problems were causing a crisis in drug development. “You can do the wrong study, use the drug in the wrong patient population, or use the wrong dose and get a false negative result, and then kill a drug. But I’m not sure I would say lack of accrual is killing potentially promising drugs,” Borghaei said.

Efforts to Enhance Enrollment

Loaiza-Bonilla said some of the studies that fail are noninferiority trials or are trying to fine-tune timing issues rather than evaluating new drugs. “Typically, a good drug will find its way, because you get good results in a patient and then you become very interested in the drug. A lot of clinical trials that fail accrual are those where the benefit is marginal.”Broader Clinical Trial Criteria

Nonetheless, it remains true that a substantial fraction of trials never complete, some studies are slow to meet their participation targets, and the majority of patients have little or no access to experimental therapies. A number of strategies have been offered to remedy these problems, including easing restrictions that automatically bar many people from participating.

In October, ASCO and Friends of Cancer Research released recommendations for broadening eligibility criteria, urging greater inclusion of children and patients with HIV, brain metastases, prior or current malignancy, or renal, liver or cardiac disease (Table).⁶ Advocates say these changes would allow more people to access new drugs and result in studies that better reflect efficacy in real-world populations.

Physicians said robust communication and education of potential trial patients can also help boost accrual. This means making time to discuss why clinical trials are important and worthwhile to patients, to explain that patients can exit trials whenever they want, and to address concerns about randomization and placebos.

“Sometimes it takes an hour to go over the rationale for a study and how a study is done, as opposed to maybe 15 or 20 minutes to talk about the standard chemotherapy and the [adverse] effects. It’s just a matter of how committed a center and a physician are to a specific study and how much time they want to spend explaining,” Borghaei said.

Another approach is to improve the process of matching and enrolling patients to avoid missed opportunities and delays that lead people to opt out. Mayo Clinic and IBM Watson Health recently piloted Watson for Clinical Trial Matching, a computing system that tracks open trials, analyzes inclusion criteria, determines a patient’s eligibility and provides trial options during office visits.⁷ A number of companies and organizations have similar initiatives. The Watson system shortened time to screening and contributed to an 80% increase in enrollment to Mayo’s systemic therapy clinical trials for breast cancer in the 11 months after implementation.

However, satisfying demand for trial participants, especially when studying narrowly targeted drugs, requires improving matching systems and reaching a larger pool of patients. “If you want to fix the system, you have to actually go where patients are,” Tolcher said. “That’s one of the reasons I left my former institution and joined Texas Oncology, which has more than 400 physicians. The potential of building a clinical trials unit that does early clinical studies, where you have a referral source from many communities in the state of Texas, is very appealing.”

Bringing trials to patients can also help improve the representation of nonwhite patients and make participation possible for people who cannot afford to travel to distant cancer centers. Borghaei said he’s seen the benefits of having a treatment facility at Temple University Hospital in North Philadelphia, which is closer to the city center than Fox Chase’s main site. “Just physically being able to open studies where those patients are seen I think can be helpful in improving minority participation. This is one of those examples where you can take a trial to where those patients are,” he said.

Turning to Technology

According to Loaiza-Bonilla, the future of trials is in sophisticated systems that store and analyze patients’ biomarker data and are paired with programs that activate clinical trials in community settings. He pointed to the example of SYNERGY-AI, a registry just launched by Massive Bio, a New York informatics provider that serves community oncologists. The system, which is itself being studied in a clinical trial, extracts data from electronic health records and matches patients to a database of studies to improve their likelihood of joining a trial.^8,9 Loaiza-Bonilla is chief medical advisor and a cofounder of Massive Bio.

Similarly, the genetic testing company Foundation Medicine has a system called FoundationSmartTrials that analyzes tens of thousands of genomic profiles and enables “Just- In-Time Trial Access,” with instant activation of sponsor trial sites close to identified patients.¹⁰

Achieving a comprehensive system for global screening and trial enrollment would require a number of complicated steps, including setting up large central data warehouses or registries, having many more patients undergo expensive genomic profiling, managing privacy issues, and making it attractive and easy for small practices to become trial sites quickly.

Such efforts are in their early stages, but Loaiza-Bonilla said they have the potential to revolutionize patient access to cutting-edge therapies. “That’s the way we’re actually going to change the treatment of cancer,” he said. “So, the future is bright, but it takes a lot of time and effort.”

References

1. Unger JM, Cook E, Tai E, Bleyer A. The role of clinical trial participation in cancer research: barriers, evidence, and strategies. Am Soc Clin Oncol Educ Book. 2016;35:185-198. doi:10.14694/EDBK_156686.
2. Bennette CS, Ramsey S D, McDermott CL, Carlson JJ, Basu A, Veenstra DL. Predicting low accrual in the National Cancer Institute’s Cooperative Group Clinical Trials. J Natl Cancer Inst. 2016;108(2):1-7. doi: 10.1093/jnci/djv324.
3. Nisen M. A hot cancer-drug race keeps getting more crowded. Bloomberg Businessweek. June 2, 2017. bloomberg.com/news/articles/ 2017-06-02/immune-oncology-drug-trials-are-getting-crowded. Published June 2, 2017. Accessed April 22, 2018.
4. Pharmaceutical Research and Manufacturers of America. List of 2017 immuno-oncology medicines in development. PhRMA website. phrma. org/report/list-of-2017-immuno-oncology-medicines-in-development. Published June 1, 2017. Accessed April 22, 2018.
5. Drilon A, Laetsch TW, Kummar S, et al. Efficacy of larotrectinib in TRK fusion—positive cancers in adults and children. N Engl J Med. 2018;378(8):731-739. doi: 10.1056/NEJMoa1714448.
6. ASCO and Friends of Cancer Research release comprehensive recommendations to broaden eligibility criteria for cancer clinical trials. American Society of Clinical Oncology website. asco.org/advocacy-policy/ asco-in-action/asco-and-friends-cancer-research-release-comprehensive. Published October 2, 2017. Accessed April 22, 2018.
7. Madson R. Mayo Clinic’s clinical trial matching project sees higher enrollment in breast cancer trials through use of artificial intelligence [news release]. Rochester, MN; Mayo Clinic; March 8, 2018. newsnetwork.mayoclinic.org/discussion/mayo-clinics-clinical-trialmatching- project-sees-higher-enrollment-in-breast-cancer-trialsthrough- use-of-artificial-intelligence/. Accessed April 22, 2018.
8. Massive Bio announces first patient enrolled in its pivotal global SYNERGY-AI: Artificial Intelligence Based Precision Oncology Clinical Trial Matching And Registry [news release]. New York, NY: Massive Bio, Inc; March 15, 2018. massbio.io/news-massive-bio-announces- first-patient-enrolled-in-its-pivotal-global-synergy-ai-artificial-intelligence- based-precision-oncology-clinical-trial-matching-and-registry. Accessed April 22, 2018.
9. SYNERGY-AI: artificial intelligence based precision oncology clinical trial matching and registry. clinicaltrials.gov/ct2/show/ NCT03452774. Updated March 2, 2018. Accessed April 22, 2018.
10. FoundationSmartTrials. Foundation Medicine, Inc website. foundationmedicine.com/insights-and-trials/foundation-smarttrials. Accessed April 22, 2018.