More than half of Australians with relapsed diffuse large B-cell lymphoma would have been excluded from every landmark trial underpinning new therapies, raising questions about how well trial results translate to routine practice.
More than half of Australian patients with relapsed diffuse large B-cell lymphoma would not have qualified for any of the landmark clinical trials that led to the approval of many of the therapies they now receive, according to new research that questions how well trial outcomes translate into routine clinical practice.
The multicentre study, published in the British Journal of Haematology, examined seven registrational studies that supported accelerated approvals for novel treatments, including CAR-T cell therapies, antibody-drug conjugates, and targeted agents.
Researchers at the Olivia Newton-John Cancer Research Institute (ONJCRI) then retrospectively applied each trial’s eligibility criteria to 180 Australian patients who experienced 320 episodes of relapsed disease between 2012 and 2022.
Overall, 52% of relapse episodes were ineligible for every trial, while no patient met the eligibility criteria for all seven studies.
Eligibility varied widely between trials, ranging from just 4% for the SADAL trial of selinexor to 22% for the ZUMA-1 trial of axicabtagene ciloleucel, reflecting marked differences in enrolment criteria rather than patient characteristics alone.
The researchers identified substantial heterogeneity across the registrational studies despite all evaluating treatments for the same disease setting.
The trials required a median of 39 eligibility criteria, with between 31 and 45 inclusion and exclusion requirements. Of 21 clinical eligibility domains assessed, only three showed complete agreement across all studies: minimum age, pregnancy status, and active hepatitis B infection.
By contrast, there was no agreement between trials on definitions of measurable disease, cardiac function requirements, lymphocyte thresholds, or serious infection criteria. Requirements for renal and liver function, previous therapies, and performance status also varied considerably.
The most common reason patients failed to qualify was the required number of previous lines of therapy, excluding between 41% and 70% of patients depending on the study. Other frequent reasons included impaired renal function, poor ECOG performance status, and active infection. Some patients became eligible for more studies with successive relapses as they accumulated additional prior treatments.
Related
“This supports the haematology community’s concerns that clinical trial cohorts and real-world cohorts are really very different,” she said.
“Only the ‘best’ patients are eligible for these clinical trials, and once the treatment has been approved by the regulatory bodies, there is far less regulation over which patients receive the treatment in the clinics. This can expose patients to unexpected toxicity and lower than expected treatment efficacy.
“We recognise that early-stage trials require robust eligibility criteria to ensure patient safety. However, registrational trials should really adopt more inclusive and permissive criteria to better reflect the characteristics of real-world populations.”
The researchers noted that clinicians routinely extrapolate efficacy and safety data from registrational studies to patients who would not have met the original eligibility criteria.
Previous post-marketing studies of therapies such as polatuzumab vedotin, tafasitamab-lenalidomide, and CAR-T therapies have consistently reported lower response rates and poorer progression-free survival than those seen in the pivotal trials.
Interestingly, the Australian cohort found no significant difference in overall survival between patients who would have been eligible for at least one registrational trial and those who were ineligible for all of them.
Progression-free survival in the real-world cohort was also broadly comparable with outcomes reported in the original trials, suggesting restrictive eligibility criteria did not simply identify patients with inherently better prognoses.
The authors said that while strict eligibility criteria remained appropriate for early-phase safety studies, registrational trials intended to support regulatory approval should adopt more inclusive and standardised criteria.
They also called for greater use of disease registries and other real-world data sources to help ensure trial populations better reflect patients likely to receive new therapies after approval.
More than 26,000 Australians are living with one of more than 80 lymphoma subtypes, making lymphoma the country’s most common blood cancer.
The researchers said clinicians should consider the eligibility criteria of the original registrational studies when interpreting efficacy and safety data and counselling patients about the likely benefits and risks of newly approved treatments.
“Registrational trials in rDLBCL employ stringent and highly variable eligibility criteria that exclude most real-world patients, substantially limiting the clinical applicability of reported outcomes,” they concluded.
“Our findings add to the growing evidence supporting the need for harmonised, less restrictive eligibility frameworks across trials within the same disease indication.
“Integrating real-world data early in drug development is critical to aligning study populations with routine clinical practice and ensuring that future therapies deliver meaningful benefit to the patients most likely to receive them.”
