OBJECTIVES: Number needed to treat (NNT) captures treatment effectiveness by indicating the number of patients necessary to receive active treatment to prevent one additional bad outcome versus a control. The main concern of NNT is finding an acceptable balance between the benefits of the intervention and the severity of the bad outcome. In this review, the ranges and averages of reported NNTs in oncology studies were examined to address this concern. This is an update of a previously published review by Azimpour et al. 2019.

METHODS: A targeted literature review was conducted in PubMed to identify oncology studies reporting NNT. Relevant records between June 6^th, 2019, and June 29^th, 2022 were extracted and assessed. Descriptive statistics were performed to identify factors influencing NNT values and help determine the positioning of new NNT values relative to recent published literature.

RESULTS: From 83 studies identified, 12 met inclusion criteria. Among included studies, 2 were in renal-cell carcinoma and 10 others were in distinct cancer types, with 7 cancer types that were not included in the prior review. Reaffirming the previous review, cancer type, patient characteristics, clinical endpoints, and length of follow-up were observed to have an impact on NNT. In general, NNTs trended upwards with increasing follow-up time. In studies stratifying by risk, lower NNTs were observed in low-risk disease than high-risk, except for one study on prostate cancer, in which the inverse was true.

CONCLUSIONS: This update includes fields and cancer types that were not captured in the previous review but reasserts its findings: factors such as cancer type, clinical endpoints and patient characteristics may influence NNTs in oncology. NNT values should not be considered naively without taking into consideration the context of the evaluation. Future research is required to explore links between these factors and NNT values.