The FDA recently released 5 new draft guidance documents that promote broader patient eligibility for cancer clinical trials.1 The policies encourage inclusion of certain individuals who were previously disqualified due to medical conditions or biological factors, including brain metastases, organ dysfunction, prior or concurrent malignancies, chronic infections, and age.
According to Scott Gottlieb, MD, former FDA commissioner, current eligibility criteria have become too restrictive for various reasons, ranging from historical precedence to intentional design. “Some eligibility criteria have become commonly accepted over time or used as a template across trials without a clear scientific or clinical rationale or justification,” Gottlieb said in a press release. “In other cases, eligibility criteria can be deliberately restrictive.”
Howard S. Hochster, MD, associate director for clinical research and chief of gastrointestinal medical oncology at Rutgers Cancer Institute and director of cancer clinical research for oncology services at RWJBarnabas Health, both in New Brunswick, New Jersey, described how particular exclusions may lead to uncertainty in routine practice, using checkpoint inhibitors as an example.
“With all the anti–PD-1 agents, every trial systematically excluded people who had any kind of autoimmune disease,” Hochster said in an interview with Targeted Therapies in Oncology(TTO), noting that even minor autoimmune-related historical events, such as a history of suspected Hashimoto’s thyroiditis, could lead to exclusion. Without relevant clinical trial data from patient populations that are encountered regularly in clinical practice, Hochster said that clinicians have had to “cross their fingers and hold their breath while they are treating people.”
Retrospective studies have filled some of the knowledge gaps, at times highlighting the unmerited reality of certain eligibility restrictions. For instance, concerning anti–PD-1 agents, a 2018 study by Giulia C. Leonardi, MD, et al showed that patients with autoimmune disease who were treated with checkpoint inhibitors had similar rates of immune-related adverse events (AEs) compared with patients without autoimmune disease. In addition, few patients with autoimmune disease had exacerbations of their preexisting condition.2
Although the results from such studies can provide valuable insights, often by drawing data from electronic medical records, retrospective research is not ideal, said Howard “Skip” Burris, III, MD, president of clinical operations and chief medical officer at Sarah Cannon Research Institute in Nashville, Tennessee, and president-elect of the American Society of Clinical Oncology (ASCO). “Frankly, it is a shame we have to do that,” Burris said in an interview with TTO. “It would be nice to have clinical trials that would do that [research] in a prospective fashion.”
Burris suggested that this ambition is shared among several other stakeholders, as demonstrated by the collaborative nature of the FDA’s initiative to influence clinical trial reform. “In 2017, ASCO and Friends of Cancer Research released a statement recommending strategies to change the exclusionary nature of eligibility criteria,” Burris said. “These consensus recommendations came from experts in the field, and stakeholders in the cancer community [who] were dealing with [these challenges] every day.”
Along with offering more real-world data, the new policies should increase clinical trial enrollment, Burris noted, leading to better outcomes in the near future. “It has been shown in a variety of settings that participation in a clinical trial is often the best option for a patient and that participating in a trial leads to better outcomes, more support, more eyes on the patient, and more adherence to the protocol and program,” he said.
At the 2019 ASCO Annual Meeting, 1 retrospective analysis conducted in non–small cell lung cancer found that these expanded criteria would enable nearly twice as many patients to consider trial participation. Out of a total of 10,346 patients eligible for trial participation under the broadened criteria, 21.2%, 21.5%, and 14.4% would have been deemed ineligible due to brain metastases, prior or concurrent malignancies, and renal function, respectively.3
When asked if the new draft guidance documents come with any drawbacks, Burris denied any; in contrast, Hochster suggested that the picture is not entirely positive.
“I think that there is a downside to this process,” he said. “It will make interpreting clinical trials a little bit more difficult.” Specifically, intercurrent illness may complicate interpretation of AEs, he said. “The more you expand the universe of patients you are treating from the ‘ideal perfect patient,’ the more you are going to have some noise in the system.”
Burris offered a different viewpoint. “The hope is that while the data might be different than what we have seen in the past, I think we will know how the trial was written, we will know the patients [who] are accrued, and the data will be the data to support it,” he said. “It should be more realistic and more consistent with what physicians should be expected to see in their practices.”
On this latter topic, Hochster agreed, emphasizing that the new guidance documents should lead to more pragmatic data (FIGURE). “I think that by broadening the universe of eligible patients to the clinical trial, we will have a better reflection of the patients we tend to see in our clinics,” he said. “In the long run, as more drugs are tested in a wider group of patients, or a less select group of patients, we will have a better sense that [the data may be applied] to our everyday patients in the clinic.”
Although the new FDA draft guidance documents, if adopted, are not explicitly enforceable, Hochster and Burris anticipate compliance, predicting that future conversations between the FDA and drug developers will be influenced by these new recommendations.
“It is going to cause more back and forth between the drug companies and the FDA,” Hochster said. “If these questions are being raised all the time, I think it will help the drug companies feel more comfortable not excluding those patients. And they will have to actively do something to make sure that they exclude them instead of going along with the flow. Most of the time, if you are [talking about a] drug company and the FDA says, ‘Do this’, then 9 out of 10 times, they are going to agree.”
“It is going to be one study at a time,” Burris said. “I am optimistic that there is going to be a gradual transition into seeing this become the new standard, and it very well may take a few years, but I think we will get there.”
“Historically, we have always excluded [patients with brain metastases] from clinical trials because the feeling was that the drugs do not work that well [in this population and their participation] would not have reflected the benefit of the drug,” Hochster said. “But once people started living longer and longer with immunotherapy, it became important to treat patients with brain metastases as well.”
To this end, the FDA recommends including patients with brain metastases in all cancer clinical trials if it is safely possible, preferably throughout the drug development process.4 The FDA emphasizes inclusion of patients with brain metastases in trials for cancers that commonly metastasize to the brain, such as lung cancer and melanoma. Baseline central nervous system (CNS) screening should be performed when brain metastases are present or likely to occur, and detection of a brain lesion should not result in automatic exclusion. If a “strong rationale” exists for exclusion of certain patients, investigators are urged to describe this in the trial protocol.
Multiple justifiable scenarios are described in the guidelines, such as the exclusion of patients who may be at an increased risk of seizures and bleeding or hemorrhage events as a result of the prescribed agent. Furthermore, patients receiving steroids that may compromise agent efficacy should be considered for exclusion. Following these general considerations, the FDA describes trial eligibility in terms of 3 forms of brain metastases: treated/stable, active, or leptomeningeal.
Patients with treated or stable brain metastases should be included regardless of an agent’s predicted ability to penetrate the blood–brain barrier. When included, patients should be neurologically stable to limit confusion of metastatic symptoms with CNS toxicity from a tested agent. Ideally, patients taking corticosteroids should be on a stable or decreasing dose in the week prior to study commencement.
Patients with active brain metastases should be included in clinical trials if CNS treatment is not immediately necessary or likely to be necessary; however, known risks of CNS toxicities may warrant exclusion.
Patients with leptomeningeal metastases should not be automatically excluded from clinical trials. Similar to patients with active brain metastases, those with leptomeningeal metastases who do not require immediate treatment for CNS disease or are unlikely to require treatment in the short term should be included.
According to Burris, the new FDA recommendations surrounding organ dysfunction represent “fairly minor shifts” compared with previous guidance. Specifically, the document calls for inclusion of patients with “relatively preserved organ function.” Further recommendations are broken down into 3 categories by organ system: renal, cardiac, and hepatic.5 Some of the key principles are summarized as follows:
1) Kidney function should be measured by glomerular filtration rate, with a consistent equation used throughout the drug development process. When applicable, prospective dose adjustments may be considered for patients with reduced kidney function to facilitate similar exposure as patients with normal kidney function.
2) Patients with “significant clinical cardiac abnormalities” should be excluded, particularly in early-phase studies. Ejection fraction should be measured by echocardiography or multigated acquisition scan, with the note that no minimum value has been established to predict development of heart failure. Concerning QTc prolongation, baseline clinical evaluation should be developed collaboratively with the FDA.
3) Because drug metabolism through the liver cannot be accurately predicted by liver biochemical testing, the FDA recommends comprehensive patient assessment. Estimations by the Child- Pugh or Model for End-Stage Liver Disease should be considered in light of patient history and epidemiological, environmental, and genetic factors.
“Impaired hepatic blood flow, in particular, when portal hypertension is evident, is critical to assessing liver handling of drugs,” the FDA guidance noted. “Therefore, even in the setting of low Child-Pugh or Model for End-Stage Liver Disease scores, normal international normalized ratio/prothrombin time, and normal total bilirubin, more reliable measures to predict biotrans formation, conjugation, and drug permeability related to hepatic transformation of xenobiotics are needed.”
Prior or Concurrent Malignancies
Burris suggested that the new recommendations for prior or concurrent malignancies are becoming increasingly relevant. “It is a fact now that patients are living much longer,” Burris said, “and we are heading toward having 20 million cancer survivors in this country. It is going to be more common, and we see it every week in our clinic.”
Now, according to the FDA, patients with “prior or concurrent second primary malignancy” should be included if “natural history or treatment does not have the potential to interfere with the safety or efficacy assessment of the investigational drug.”5
The FDA specifically discusses HIV, hepatitis B virus, and hepatitis C virus in its guidance document for patients with chronic infections.6 Inclusion is encouraged when patients infected with these viruses have adequate immune system and organ function. According to Burris, the recommendations are fitting for modern practice. “We have great therapies now that are successful in treating HIV and hepatitis, so this is making [clinical trial data] more timely and consistent with the results we are seeing today.”
The final 2 draft guidance documents outline eligibility criteria for children (2 to <12 years) and adolescents (≥12 to 17 years).7,8 In general, inclusion may be considered if cancer type or molecular target is shared between younger and adult patients, potential benefit could justify risk of participation, and no curative options are available to the patient. Dose modifications may be considered, if necessary, and long-term prospective follow-up is encouraged, particularly if treatment could impact growth and development or result in late effects.