Expanding Access to Cancer Clinical Trials for Patients With Mental Illness

Depriving patient populations of access to trials that can establish new standards of cancer care without sound rationale creates unjust inequities in cancer care. Thus, ASCO and the Friends of Cancer Research recently issued a persuasive call to action to broaden clinical trial eligibility criteria so that oncology clinical trials are more representative. To promote this expansion, the ASCO recommendations included suggested language to eliminate unnecessarily restrictive inclusion and exclusion criteria in the following five areas: minimum age requirements for trial enrollment, HIV/AIDS status, brain metastases, organ dysfunction, and prior and concurrent malignancies.^1,2 To increase access to trials and inform oncology practice for the diverse populations we serve, we need to further extend this paradigm shift to other underrepresented groups.

In this commentary, we recommend extending this paradigm of inclusiveness to mental illness. Individuals with serious mental illness (SMI), particularly bipolar disorder and schizophrenia, have two to four times greater mortality from breast, lung, colorectal, and oral cancers compared with adults without mental illness.^3-7 Furthermore, adults with SMI die 15 to 30 years earlier than the general population, and cancer is the second leading cause of death. Inequities in cancer care may lead to premature mortality.⁸ Barriers to trial participation may contribute to this disparity. Although mental illness (ranging from adjustment disorders to anxiety, mood, and psychotic disorders) affects between 20% and 38% of patients with cancer and is associated with poor cancer outcomes, patients with psychiatric disorders are frequently excluded from cancer clinical trials.⁹

Psychiatrists practicing in academic cancer centers, medical oncologists who lead clinical trials, and disparities researchers grapple with whether it is ethical to include patients with mental illness in clinical trials. For example, can a patient with bipolar disorder provide informed consent and adhere with the complex monitoring protocol of a novel therapy for advanced laryngeal cancer? Will a patient with severe anxiety and advanced lung cancer complete all required monitoring? As clinicians and researchers, we are attuned to our responsibility to protect vulnerable populations from coercion, maintain patient safety, and promote the integrity of the clinical trial. However, we rarely ask whether it is ethical to not include patients with mental illness in clinical trials. When clinical trials offer access to novel therapies to patients who have exhausted standard treatments, access to trials becomes a human rights issue and a fundamental component of cancer care equity. From a social justice perspective, overly restrictive eligibility criteria may prevent an individual patient with a psychiatric disorder from accessing a potentially life-saving treatment. Furthermore, by excluding patients with mental illness from trials, we miss the opportunity to tailor cancer care delivery for an understudied population and therefore perpetuate disparities in cancer outcomes.

Individuals with mental illness are commonly excluded from clinical trials; the specific rationale for exclusion is rarely explicated. One half of highly cited randomized trials across medical specialties excluded patients for psychiatric reasons.⁹ Furthermore, two thirds of Investigational New Drug applications filed with the US Food and Drug Administration (FDA) excluded patients with psychiatric illness, social situations limiting compliance, or substance abuse.¹⁰ Beyond the group of patients who are explicitly excluded from clinical trials, clinicians can exercise significant discretion when evaluating whether to approach patients to participate. To understand inequities in access to clinical trials for people with mental illness, we need to assess protocol language (eg, ClinicalTrials.gov), published inclusion and exclusion criteria, and researcher practices regarding approaching and enrolling patients with mental illness. If one half of clinical trials exclude individuals with mental illness, we will continue to lack data to inform evidence-based cancer care targeting this population.

Consider the example of varenicline, the most recent and highly effective tobacco cessation pharmacologic treatment. Although nearly 50% of individuals with SMI smoke tobacco,¹¹ the original varenicline trials excluded patients with mental illness, so the safety of the medication for patients with mental illness was unclear. Then, during postmarketing surveillance, reports emerged about suicidal ideation and worsening depression, leading to an FDA black box warning and a fear of prescribing this medication to patients with mental illness, so these patients did not use this new and improved medication to quit smoking.

However, when the FDA commissioned a randomized trial including patients with stable, treated mood and psychotic disorders and Pfizer (New York, NY) supported conducting a large randomized trial, it was shown that varenicline was not associated with an increase in neuropsychiatric adverse effects and that patients with SMI who received varenicline were four times more likely to stop smoking.¹² Subsequently, the black box warning was removed, yet rates of varenicline use among patients with mental illness remain lower than the general population.¹³ By shying away from prescribing varenicline to patients with mental illness, clinicians may miss the opportunity to address and treat a preventable cause of mortality.

Recently, our clinical research team conducted qualitative interviews with oncology and mental health clinicians and elicited their beliefs about clinical trial participation.¹⁴ Many clinicians shared their reticence to approach patients with mental illness to participate in clinical trials. Oncology clinicians shared their concerns about enrolling patients with mental illness, which included the inability to provide informed consent, lack of motivation to participate in cancer research, and inability to complete trial procedures or return for monitoring appointments. In addition, oncologists worried that including patients with schizophrenia could increase burden for patients with complex needs, adversely affect their safety, and increase the time and risk for investigators and the trial itself.

Although these clinicians’ concerns are sometimes justifiable, there are research-supported solutions to facilitate clinical trial participation for patients with SMI. Addressing the issue of informed consent, multiple trials provide evidence that adults with mental illness can consent to clinical trials.¹⁵ Patients with mental illness, and specifically schizophrenia, have a heterogeneity of cognitive functioning and levels of insight, which can create complexity and concern for consenting investigators. However, with modified trial procedures, patients with psychotic disorders can improve their ability to understand the risks and benefits of the trial and provide informed consent to participate in most trials.^16,17 Ethically, it is important to understand patient perspectives on trial participation. Compared with patients with medical illness, patients with mood disorders are equally likely to express interest in clinical trial participation and identify similar reasons for participating in clinical trials.¹⁸ Regarding risk of patient harm, a recent review of National Institute of Mental Health data has demonstrated the safety of patients with treatment-resistant mood disorders participating in clinical trials that included psychiatric medication adjustment and placebo arms.¹⁵

In terms of patient trial access and participation adherence, new approaches have demonstrated promise in the effort to identify, enroll, and retain patients with mental illness in clinical trials including community partnerships, modified eligibility criteria and consent processes, and person-centered engagement. Our team is building a community network of researchers, clinicians, patients, and advocates dedicated to increasing access to research for individuals with mental illness. Rigorous evaluation plans are needed to assess the impact of this effort. Patient and caregiver stakeholders have echoed the desire to partner in research initiatives and provided feedback about decreasing barriers to trial participation, including decreasing mental health stigma, recruiting from community mental health clinics, and collaborating with residential front-line staff.¹⁹ Although data are limited regarding follow-up rates for patients with SMI in cancer clinical trials, targeted outreach has increased recruitment and retention of patients with SMI in clinical trials. For example, a recent randomized trial of patients with SMI and diabetes that addressed barriers to trial participation, including stigma and lack of transportation, achieved 75% retention at 5 years.²⁰

Park et al²¹ recently completed a randomized trial of tobacco cessation integrated with cancer care. Rather than exclude all patients with mental illness, investigators excluded only patients with uncontrolled psychosis, active suicidal ideation, or psychiatric hospitalization within the past year. Importantly, rates of consent, enrollment, and adverse events for patients with mental illness were equivalent to other trial participants.²² Irwin et al²³ conducted a trial of a psychosocial intervention for patients with pre-existing mental illness and cancer and developed trial procedures that addressed multilevel barriers to trial participation. For example, study clinicians used a verbal consent process with concrete language to decrease paranoia and encourage conversation about the risks and benefits of the trial. The enhanced consent procedures were informed by the MacArthur Competence Assessment Tool for Clinical Research and have been shown to increase patient understanding of the purpose, risks, and benefits of trials through repetition and dialogue.^24-27 Recognizing that patients with SMI experience disproportionate poverty and social isolation, the team used a person-centered approach, including flexibility with visit timing and location, to increase retention and completion of assessments without adverse events.²³

In addition, integrated models of cancer and mental health care delivery may make it possible to treat the psychiatric illness and therefore improve the ability of patients to complete trial procedures.^28,29 Alternatively, patients’ guardians may be able to provide informed consent with patients providing assent. By considering mental illness like other comorbid conditions, such as HIV (also complex and stigmatized), researchers can adopt a more nuanced approach and include patients they may have previously excluded based on diagnosis or templated language from previous trials.

As the ASCO statements underscore, exclusion criteria need to be related to the toxicity of the treatment being investigated. For example, rather than excluding all patients with kidney disease when trial drugs are renally cleared, experts recommend that we assess creatinine clearance. Rather than excluding all older adults, we consider medical comorbidities, drug interactions, and functioning. However, when determining eligibility criteria, we rarely specify the psychiatric diagnosis, severity of symptoms, amenability to treatment, and risks and benefits of the therapy being investigated. Instead of excluding all patients with mental illness, we can evaluate whether the trial has anticipated effects on cognition or mood and consider the severity of psychiatric symptoms and amenability to treatment. In addition, we need to consider the risks, benefits, and level of uncertainty of the proposed trial. For example, trial participants need a greater complexity of understanding to consent to a phase I trial with a high risk of toxicity and low likelihood of individual benefit compared with a phase III trial of an approved immunotherapy with low risk of toxicity, greater likelihood of individual benefit, and limited alternatives to treat an aggressive cancer.

Altogether, we must balance the protection of research participants with respect for patient autonomy, consideration of potential benefit from the trial, and social justice implications. Acknowledging the history of mistreatment of individuals with mental illness in clinical research,³⁰ it is important to be thoughtful about protection of vulnerable populations; however, the pendulum seems to have swung too far. Restrictive eligibility criteria exclude patients with mental illness from potentially life-saving research and treatment and limit research advances that could ameliorate cancer disparities. Therefore, we strongly recommend that serious consideration be given to extending work to broaden trial eligibility criteria for patients with mental illness. In Table 1, we suggest engagement strategies and recommended language to promote inclusiveness in access to clinical trials. We must recognize the unintended consequences of excluding all patients with mental illness instead of designing eligibility criteria that reflect a compelling rationale for exclusion that outweighs potential benefit. Because engaging under-represented populations can be time consuming,³¹ we need to support investigators in their efforts to include high-risk populations and dedicate time to engaging the patient’s clinical team and community supports. In addition, to inform future studies and target accrual, it would be beneficial to track rates of study completion and missing data separately in patients with mental illness. As an oncology community, we have an opportunity to promote equity in clinical trial participation for individuals with SMI. Given our mission to close the cancer mortality gap, we have a moral imperative to gather generalizable evidence that can decrease disparities in cancer outcomes for a population facing significant barriers to care.

TABLE 1.

Addressing Barriers to Trial Participation for Patients with Mental Illness

AUTHOR CONTRIBUTIONS

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Expanding Access to Cancer Clinical Trials for Patients With Mental Illness

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/site/ifc.

Beverly Moy

Consulting or Advisory Role: Motus (I), Remedy Partners (I), Dark Canyon (I)

Research Funding: Puma Biotechnology (Inst)

Lauren E. Fields

Employment: Janssen of Johnson & Johnson (I)

Stock and Other Ownership Interests: Janssen of Johnson & Johnson (I)

Research Funding: Janssen of Johnson & Johnson (I)

Patents, Royalties, Other Intellectual Property: US Patent No. 6586250: Method of polymerase chain reaction–based gene targeting; as employee of Washington University School of Medicine (I)

Travel, Accommodations, Expenses: Janssen of Johnson & Johnson (I)

Elyse R. Park

Honoraria: UpToDate

Research Funding: Pfizer

Lori Wirth

Consulting or Advisory Role: Merck, Loxo, Blueprint Medicines, Eisai, Amgen, Novartis, Ayala Pharmaceuticals, Cue Biopharma

Other Relationship: Iovance Biotherapeutics

No other potential conflicts of interest were reported.