Clinical trial protocols often incorporate strict eligibility criteria to maximize reproducible, results with minimal risk to trial participants. Unfortunately, restricting a trial’s patient population may result in outcomes that are not generalizable to “real-world” patients, according to a report published in Clinical Lymphoma, Myeloma & Leukemia.
In the Connect MM Registry study, Jatin J. Shah, MD, of the Department of Lymphoma/Myeloma at the MD Anderson Cancer Center in Houston, Texas, and co-authors reviewed eligibility criteria of recent randomized, controlled trials (RCTs) of patients with newly diagnosed myeloma to determine whether the criteria were too strict, and whether the populations enrolled accurately reflected their real-world counterparts.
The multicenter, prospective, observational cohort study included 1,406 adult patients (median age = 67 years; range = 24-94 years) who satisfied the minimum eligibility requirements to participate in an RCT (i.e., meeting calcium elevation, renal insufficiency, anemia, and lytic bone lesion [CRAB] criteria). All patients had symptomatic disease and were diagnosed within two months of enrollment (between September 28, 2009, and December 14, 2011). Patients were treated according to standard practice at each of the 250 study sites.
To identify common eligibility criteria, the authors searched PubMed for phase III myeloma clinical trials conducted in the past 10 years, including studies of thalidomide, lenalidomide, and bortezomib.
Studies of relapsed/refractory disease, studies of supportive care treatments, and secondary reports were excluded. The authors found 24 RCTs, all of which had the following exclusion criteria:
- absence of measurable disease
- grade >2 peripheral neuropathy
- Eastern Cooperative Oncology Group (ECOG) performance status score of 3 or 4
- a history of myelodysplastic syndromes, other hematologic malignancies, or solid tumors
Applying these common criteria to the unselected Connect MM Registry, 563 (40%) of the patients would be considered ineligible for RCTs. If exclusion criteria were more aggressive, to include M-protein (≤1.0 g/dL) and hemoglobin (≤8 g/dL) criteria, as many as 56.8 percent of patients would be ineligible for inclusion.
Most patients received a bortezomib-containing regimen as initial therapy in both the RCT-ineligible and -eligible cohorts (72.5% vs. 70.7%; p=0.842), but RCT-ineligible patients were less likely to receive lenalidomide (40.5% vs. 49.7%; p<0.001). Also, fewer RCT-ineligible patients received autologous hematopoietic cell transplantation (AHCT; 28.4% vs. 38.0%; p<0.001).
The researchers then compared patient characteristics between those who were RCT-ineligible (n=563) and those who were RCT-eligible (n=843). Significantly more RCT-ineligible patients had hypercalcemia (11.0% vs. 5.5%; p<0.001), creatinine levels >2.0 mg/dL (38.9% vs. 6.2%; p<0.001), and low hemoglobin levels (59.5% vs. 39.5%; p<0.001).
At data cut-off (January 7, 2016), the median follow-up of patients enrolled in the registry was 39.3 months (range not provided). The three-year overall survival (OS) rate appeared to be lower in RCT-ineligible patients, compared with RCT-eligible patients (63% vs. 70%; p<0.05). However, undergoing AHCT seemed to equalize three-year OS rates, regardless of RCT eligibility (81% for ineligible and 78% for eligible patients). Among patients who did not undergo AHCT, three-year OS rates remained higher for RCT-eligible patients (66% vs. 54%).
Three-year OS rates also were higher in RCT-ineligible patients who received novel therapies (including bortezomib, carfilzomib, lenalidomide, or pomalidomide), compared with RCT-eligible patients (63% vs. 70%); rates were similar among patients who did not receive novel therapy (59% vs. 80% percent, respectively).
In univariate analyses, the authors determined that the following factors were associated with increased risk of death and, therefore, could potentially serve as more meaningful exclusion criteria:
- platelet count ≤1.5×109/L
- creatinine level >2.5 mg/dL
- previous invasive malignancy
- International Staging System (ISS; I-II vs. III)
- ECOG score (3 or 4 vs. <3)
Except for previous invasive malignancy and ISS stage, the factors also were statistically significant in a multivariate analysis, with a 2.08-fold increase in the risk of death related to low platelet count (p<0.001), a 1.49-fold increase for patients with elevated creatinine levels (p<0.001), and a 1.86-fold increase for ECOG performance status score of 3 or 4 (p=0.007; see TABLE 3). “The multivariate analysis of OS and eligibility criteria reinforces the premise that RCT eligibility criteria deselect patients with more advanced disease,” the authors wrote.
Serious adverse events (AEs) were reported in a similar proportion of RCT-ineligible (66.8%) and -eligible (62.5%) patients. Infections occurred in 29.3 percent and 23.1 percent of patients, respectively, and pneumonia occurred in 13.1 percent and 12.5 percent, respectively. Serious AEs that led to hospitalization occurred in 59.7 percent of RCT-ineligible and 55.6 percent of RCT-eligible patients. Multivariate analyses identified hemoglobin <8 g/dL, creatinine level >2.5 mg/dL, previous venous thromboembolism, previous malignancy, ISS stage, and ECOG performance status as being associated with greater risk of a serious AE (see TABLE 4).
“Safety is a common endpoint of RCTs, and the perceived increased safety risks of clinical trials have been identified as a barrier to participation,” the researchers noted. “However, results of this analysis suggest that patients with newly diagnosed myeloma who are RCT-ineligible are no more likely to experience serious AEs or become hospitalized than RCT-eligible patients.”
These data identify specific criteria and parameters that have no effect on survival or serious AEs, the authors observed, including absolute neutrophil count level ≤1.5×109/L, elevated aspartate or alanine aminotransferase, hemoglobin ≤8 g/dL, and previous HIV or hepatitis infection.
“Re-evaluation of RCT eligibility criteria is needed to improve enrollment without compromising the ability of trials to determine treatment effects,” the authors concluded. “These parameters could potentially be eliminated from RCT eligibility criteria, [which] would lead to a broader patient population and accelerate enrollment without affecting the integrity of a trial or safety of patients.”
The study is limited by its inclusion of only newly diagnosed patients, which is not generalizable to the relapsed/refractory, real-world population. Also, “this analysis was able to show an association between certain eligibility criteria and serious AEs, but causality cannot be established,” the authors wrote. Future studies should examine outcomes related to race, as the present study found that 53 percent of the African-American subpopulation were deemed RCT-ineligible, compared with 39 percent in the white subpopulation (p<0.001).
Celgene Corporation provided funding for this study.
Contributing authors report financial support from Celgene Corporation, Millennium Pharmaceuticals, Onyx Pharmaceuticals, Novartis, and Array Biopharma.
Shah JJ, Bonjour R, Lazaretto C, et al. Analysis of common eligibility criteria of randomized controlled trials in newly diagnosed multiple myeloma patients and extrapolating outcomes. Clin Lymphoma Myeloma Leuk. 2017;17:575-83.