Barriers to Investigator-Initiated Clinical Trial Enrollment in Frontline Large B-Cell Lymphoma: A Single-Center Retrospective Analysis

Background:

Despite treatment advances, first-line therapy fails to cure about one-quarter of patients with newly diagnosed large B cell lymphoma (LBCL). Clinical trials are crucial for improving outcomes and survival, but accrual in 1^st line LBCL is often suboptimal due to demographic, clinical barriers, and disease-specific issues like need for diagnostic workup in the context of rapid progression and hospitalization in the highest-risk patients. While industry-sponsored trials, such as POLARIX and PHOENIX, accrue on average 2-3 patients per site annually, local investigator-initiated trials (IITs) at our center accrue 5-10 times faster, in part due to improved feasibility and design. We aimed to identify barriers to clinical trial accrual in 1^st line LBCL trials at our center to further optimize design for efficient and diverse patient enrollment.

Methods:

We identified newly diagnosed LBCL patients referred to Fred Hutch Cancer Center (FHCC), almost all of whom were pre-screened for clinical trials by a nurse navigator and study investigators between October 2022 and June 2024. Two IITs were open for frontline LBCL or high-grade B-cell lymphoma (NCT05455697, NCT04231877). We analyzed demographic, geographic, and socioeconomic factors, including sex, race, ethnicity, distance from FHCC, area deprivation index (ADI) [higher scores indicating higher neighborhood deprivation levels], insurance type, and interpreter need. We compared characteristics of patients who enrolled in the trial to those eligible but not enrolled. For ineligible patients, we identified factors contributing to ineligibility. We performed Wilcoxon rank-sum and chi-squared tests for continuous and categorical variables, respectively, comparing patients who enrolled and those who were eligible but did not enroll.

Results:

A total of 153 patients with newly diagnosed LBCL were identified and reviewed. Of these, 61% were male, 87% white, 97% non-Hispanic or Latinx. Median age was 66 (IQR 55-73). Insurance included 43% private, 7% Medicaid, 49% Medicare, and 1% other government. Four percent needed a language interpreter. Sixty-nine patients (45%) were deemed potentially trial-eligible by pre-screening; 24 (35% of those potentially eligible) proceeded to enroll in a trial. Enrolled patients lived significantly closer to FHCC (median 15 miles, IQR 7-18) than non-enrolled eligible patients (50 miles, IQR 19-109, p=0.00015). Enrolled patients also had a lower ADI (median 8, IQR 4-14) compared to non-enrolled eligible patients (median 21, IQR 8-28, p=0.006). Age, sex, race, ethnicity, need for an interpreter, and insurance type were not significantly different between the enrolled and non-enrolled eligible groups.

Among those trial-eligible but not-enrolled, reasons included patient preference due to geography (27%), need for urgent treatment (10%), and a trial enrollment pause (5%). The remaining patients (58%) received standard-of-care (SOC) based on unspecified patient or provider preferences, with 67% receiving SOC at FHCC and 33% at community practices.

Among 84 trial-ineligible patients, the median age was 68 (range 34-90), 58% male, 88% white, and 96% non-Hispanic or Latinx. Median distance from FHCC was 79 miles (IQR 24-166), and median ADI was 25 (IQR 14-39). Insurance was 40% private, 7% Medicaid, 51% Medicare, and 1% other government insurance. The most common reason for ineligibility was having already started treatment before consultation (53%), due to urgent treatment needs (15%) or starting treatment in the community before a second opinion. Other reasons included no measurable disease (10%), CNS involvement (4%), another malignancy (4%), inadequate performance status (4%), cardiac dysfunction (2%), liver dysfunction (2%), or prior doxorubicin exposure (2%).

Conclusion:

Geographic proximity and socioeconomic factors significantly influence trial enrollment. Proximity to the trial site and lower deprivation were associated with higher enrollment. No significant differences were observed in age, sex, race, ethnicity, need for an interpreter, or insurance type between enrolled and non-enrolled groups. Strategies to address these barriers can lead to more efficient, inclusive, and representative research. Additionally, expedited referral pathways are needed to improve trial access for interested patients.