Outcomes of Treatment With Cancer Biosimilars Found Equal to Those of the Reference Drug in Harvard Systematic Review and Meta-Analysis

Results of rigorous clinical evaluation of cancer drug biosimilars were statistically indistinguishable when compared with results of evaluations of the reference drug across subgroups of drugs, cancer types, and outcome measures, according to the findings of a systematic review and meta-analysis by Harvard Medical School researchers (JAMA Oncol. 2022;8(4):537-545. doi:10.1001/jamaoncol.2021.7230).

“The extent of the preapproval clinical testing that is needed and how these biosimilars compare with the originator biologic products remain critical issues in establishing a vibrant biosimilar market,” wrote Aaron Kesselheim, MD, JD, MPH., Program on Regulation, Therapeutics, and Law, Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA, and colleagues. Moreover, biologics account for almost half of U.S. drug spending, according to the study report authors.

To analyze and compare the design of efficacy studies of cancer biosimilars with the design of pivotal trials of their reference drug and to estimate summary risk ratios for each cancer type and drug subgroup, the Harvard researchers did a systematic search of published articles and abstracts by using Embase, PubMed/MEDLINE, and ClinicalTrials.gov. The search was last updated on April 18, 2021. 

From these sources, they selected reports or abstracts in English of studies that compared a disease-modifying cancer biologic drug with its biosimilar, including efficacy or surrogate efficacy results. They then extracted outcome estimates and study characteristics from each study and used random-effects meta-analyses for each molecule, cancer type, and outcome subgroup to calculate pooled relative estimates and 95% confidence intervals (CIs).

Using these methods, they compared study characteristics, including population size, blinding, and randomization, of the biosimilar trials and those of their reference drugs. They also collected risk ratio estimates for the relative change in surrogate measures such as progression-free survival for the biosimilar drug and reference drugs.

The meta-analysis included 31 studies of cancer biosimilars of 3 reference drugs. Together, these studies included more than 12,000 patients. In all 7 subgroups used, the biosimilars analyzed produced surrogate efficacy results indistinguishable from those for their reference drug. The meta-analysis also included 6 reference drug trials that together included 1811 patients. 

In general, patients in the biosimilar group outnumbered those in  the reference drug trials. For example, the mean number of patients in the biosimilar studies was 397 and 302 in the reference drug trials. Moreover, the biosimilar studies were more likely than the reference drug trials to be randomized clinical trials (RCTs) than single-group or observational studies and were more likely to be double-blinded than open-label. For instance, 100% of the 31 biosimilar studies were RCTs, compared with 50% (3 of 6) of the reference drug trials, and 84% (26 of 31) of the biosimilar studies were double-blinded compared with 17% (1 of 6) of the reference drug trials.

On the basis of these findings, Dr. Kesselheim and team concluded that “These results should help overcome skepticism that physicians and patients may have in using more cost-effective biosimilar drugs currently available in the US to manage cancer diagnoses.”