Melanoma Clinical Trial Inclusion Criteria May Need Reframing

Janice M. Mehnert, MD, of the Rutgers Cancer Institute of New Jersey, gave a presentation at the 2015 ASCO Annual Meeting entitled, “Should Comorbidities Be an Automatic Exclusion for Clinical Trials?”

Janice M. Mehnert, MD

As a phase I investigator, Janice M. Mehnert, MD, has spent her career thinking outside the proverbial box. For the 2015 ASCO Annual Meeting, she turned her attention to two key issues: (1) oncologists’ roles in leading melanoma patients to enroll in clinical trials and, more importantly, (2) deciding which patients to exclude to ensure safety and the trial’s most accurate results.

Mehnert, director of developmental therapeutics/phase I in melanoma and soft tissue oncology at the Rutgers Cancer Institute of New Jersey, gave a presentation entitled, “Should Comorbidities Be an Automatic Exclusion for Clinical Trials?” In her talk, she encouraged colleagues to stay abreast of current clinical trials and to discuss them with their patients, even those who may seem to be unlikely candidates for enrollment. “We owe it to our patients to thoroughly discuss clinical trials with them at their initial consults,” she said. “Even melanoma patients with comorbidities should receive expert consultation regarding the risks and benefits of available clinical trials.”

According to Mehnert, that need remains strong, even though standards of care now exist for melanoma patients. “Clinical trials provide an important opportunity to learn proper dose and sequencing,” she said in an interview. “And we need to know which patients will get the dramatic results that are possible with immunotherapy.”

Other reasons for patients to participate in trials include creating options for those who are not eligible for the existing standard of care, improving the assessment of existing standards and, of course, helping in the continuing quest to evaluate new therapies.

But what about patients who do not enroll in a trial? Mehnert’s curiosity was piqued by the 2014 study by St. Germain et al in theJournal of Oncology Practicethat examined factors in cancer clinical trial accrual.1“They found that there are many reasons patients don’t enroll in trials but by far the biggest reason was comorbidities or provider perception of comorbidities, disability, and medical frailty,” Mehnert said. “These findings underscore the need for providers to look carefully at each patient to better assess their fitness for trial enrollment.”

Despite the difficulties associated with these comorbidities, broader trial eligibility is necessary to allow investigators to ask the most critical research questions, Mehnert contended. “With effective standards, the trial population will inevitably narrow. Overly stringent criteria hinder the applicability of treatment to the broader population,” she said, noting that both undertreatment and overtreatment may result.

Mehnert was intrigued by the 2010 study of Chao et al in theJournal of Investigative Medicinethat reviewed eligibility from 24 trials.2After creating a list of common exclusion criteria, the researchers found that only slightly more than half the patients would have been eligible for phase III, randomized, controlled trials. The study inspired her to conduct a similar analysis of the protocols of 13 trials under way at the Rutgers Cancer Institute of New Jersey, using these comorbidities:

  • Organ dysfunction
  • Chronic infections
  • Autoimmunity/any steroid use
  • Brain metastases (metastatic protocols)
  • Second malignancies (metastatic protocols)
  • Performance Status (PS) 0-1

The results ranged from 8% for age, meaning 1/13 trials treated patient age as an exclusionary factor and 100% (13/13) excluded patients for organ dysfunction. Other results ranged from 50% (5/10) for brain metastases and second malignancies to 92% (12/13) for chronic infections.

Mehnert then considered each comorbidity. For organ dysfunction, she urged colleagues to avoid creating blanket criteria and to consider organ dysfunction-specific cohorts or trials. She also pointed to an increasing number of published case reports as new therapies are approved and come to market.3

About chronic infections, Mehnert reminded colleagues that benefits are being realized from multicenter studies in patients with cancer who also have persistent viral loads, lower CD4+ T-cell counts (HIV), and hepatitis B/C infections.

Postmarketing case studies can be a valuable information source, and clinical protocols may best capture safety data. Mehnert also urged widespread consideration of the U13 Report by Hurria et al on aging and clinical research4and their definition of an extended trial, which encourages investigators in phase III trials to reopen a relevant study arm. “That would be challenging but not impossible, and I think we need to consider doing that more often,” she said.

Autoimmunity is an exclusion in almost all protocols. “Autoimmunity is actually a spectrum,” Mehnert said, noting that low-dose, chronic prednisone use is occasionally allowed. According to Mehnert, autoimmunity may offer the most fruitful ground for re-evaluation, pointing to examples like one from her own institution that showed that, even while requiring colectomy for perforation, a relatively young patient received clear clinical benefit from ipilimumab.

Mehnert also cited examples of colleagues safely using immunotherapeutic agents in transplant patients with melanoma to bolster her call for re-evaluation of autoimmunity as a near-automatic exclusion.5,6“I’m also quite interested in the ongoing study by my colleagues Koyfman and Gastman at the Cleveland Clinic, who are looking at pembrolizumab after surgery and radiation for high-risk squamous cell carcinoma of the skin in immunocompetent and immunosuppressed patients,” she said. “It will be exciting to see those data mature.”

The most progress has been made in getting oncologists to believe advanced patient age should not be an automatic exclusion, according to Mehnert. “Age is less of a protocol-specific exclusion because we all know by now that chronological age does not automatically equal physiologic age,” she said. “However, further investigations are needed to define the role of advancing age in new therapies. There’s a perception that checkpoint inhibitors are too toxic for older patients, but that hasn’t been my experience. We just need more data.” Possible ways to get those data, again going back to Hurria’s U13 conference recommendations, could include doing an extended stage trial, stratifying for a cohort that’s age restricted, or a one-arm pilot trial.

Mehnert urged fellow oncologists to consider brain metastases and second malignancies carefully when determining whether or not a patient should be excluded from a particular trial. In the case of brain metastases, the emergence of central nervous system-specific trials and more permissive entry of patients who have had treated metastases are significant developments.

With second malignancies, Mehnert suggested that colleagues carefully evaluate the disease, the individual patient’s circumstances, and the true potential for interference with the trial’s parameters. For example, she suggested, consider enrolling patients with early-stage cancers that are statistically unlikely to harm the patient within a 3-to-5-year period. Oncologists can also consider enrolling patients with active malignancies if their treatment will not interfere with the study’s endpoint and there is no concomitant active therapy.

Similarly, physicians should be willing to reframe their assessment of a given patient as ‘ineligible’ to ‘likely to be ineligible’ until shown otherwise, Mehnert said. She would like to see a widespread reassessment of theEastern Cooperative Oncology Group(ECOG)/World Health Organization (WHO) PS score of 2.

“Nearly all clinical trials now require a PS of 0 or 1, even though as recently as five years ago including patients with a PS of 2 was fairly common,” she said. “If we exclude all the PS 2 patients from trials, how applicable will that therapeutic agent be in a real-world situation? The definition [of PS 2] is actually pretty reasonable for inclusion: Ambulatory and capable of all self-care but unable to carry out any work activities—up and about more than 50% of waking hours.”

Mehnert stressed that she is not suggesting that every trial should be open to every patient. “We’re fortunate in melanoma that we have a standard of care. Now, we need to be as inclusive as possible in our clinical trials without sacrificing patient safety,” she said. “And we have a duty to the drug as well—we should think carefully about what we need to do to get the most accurate results possible. Ultimately, I want to move our default question from, ‘Why should I do a clinical trial?’ to ‘Why not do a clinical trial?’”


  1. St Germain D, Denicoff AM, Dimond EP, et al. Use of the National Cancer Institute Community Cancer Centers Program Screening and Accrual Log to address cancer clinical trial accrual.J Oncol Pract. 2014;10(2):e73-e80.
  2. Chao HH, Mayer T, Concato J, Rose MG, Uchio E, Kelly WK. Prostate cancer, comorbidity, and participation in randomized controlled trials of therapy.J Investig Med.2010;58(3):566-568.
  3. Cavalcante L, Amin A, Lutzky J. Ipilimumab was safe and effective in two patients with metastatic melanoma and end-stage renal disease.Cancer Manag Res. 2015;19;7:47-50.
  4. Hurria A, Dale W, Mooney M, et al. Designing therapeutic clinical trials for older and frail adults with cancer: U13 conference recommendations.J Clin Oncol.2014;32(24):2587-2594.
  5. Lipson EJ, Bodell MA, Kraus ES, Sharfman WH. Successful administration of ipilimumab to two kidney transplantation patients with metastatic melanoma.J Clin Oncol.2014;32(19):e69-e71.
  6. Cecchini M, Sznol M, Seropian S. Immune therapy of metastatic melanoma developing after allogeneic bone marrow transplant.J Immunother Cancer.2015;3:10.