Background: Single arm studies pose some challenges to determining comparative effectiveness of treatments. Such is the case with the tumor agnostic tropomyosin receptor kinase (TRK) inhibitors, larotrectinib and entrectinib. Larotrectinib and entrectinib are indicated for treatment of advanced cancer in patients (pts) with NTRK gene fusions. Lacking head-to-head trials, modeling methods are useful to estimate relative benefit. A direct naïve comparison, where single arms from separate trials are compared, can be used. Here, we estimate and compare long term benefit for pts receiving larotrectinib or entrectinib for NTRK gene fusion-positive non-small cell lung cancer (NSCLC) and colorectal cancer (CRC) by extrapolating survival curves to generate expected life-years (LYs) and quality-adjusted life-years (QALYs). Methods: We used partitioned survival methods to forecast comparative effectiveness. Larotrectinib survival data were derived from an updated July 2021 analysis of 21 adults (≥18 years of age) with NSCLC and 15 pts with CRC (NCT02122913, NCT02637687, and NCT02576431). Survival data for entrectinib were derived from published literature. Extrapolation of results were necessary to estimate long-term outcomes since the mean follow-up was less than three years. Progression free survival and overall survival for both treatments were estimated using several parametric survival distributions (Exponential, Weibull, Log-logistic, and Log-normal), then selecting based on AIC and clinical plausibility. QALYs were estimated by adjusting the time spent in the pre progression and post progression health states by utility values from the literature. A discount rate of 3% was applied to LYs and QALYs. Model uncertainty was evaluated using one-way and probabilistic sensitivity analyses. Results: In the larotrectinib trials, response rates were 86% and 33% in NSCLC and CRC, respectively. Exponential curves were fit for both treatments. In the entrectinib trials, response rates were 70% in NSCLC and 29% in CRC. The larotrectinib model estimated 7.3 LYs (95% Credible Interval [CrI]: 3.3, 11.4) and 4.3 QALYs (95% CrI: 2.2, 6.5) in NSCLC and 1.7 LYs (95% CrI: 0.8, 3.2) and 1.3 QALYs (95% CrI: 0.7, 2.3) in CRC. The entrectinib model estimated 2.7 LYs (95% CrI: 1.7, 4.2) and 1.6 QALYs (95% CrI: 1.1, 2.3) in NSCLC and 0.5 LYs (95% CrI: 0.2, 1.6) and 0.4 QALYs (95% CrI: 0.1, 1.1). Compared to entrectinib, larotrectinib produced additional gains of 4.7 LYs and 2.7 QALYs in NSCLC and 1.1 LYs and 0.9 QALYs in CRC. Limitations include non-matched baseline differences in the trial populations. Conclusions: In the absence of head-to-head trials, comparative effectiveness can be assessed by extrapolating from single armed studies indicated for the same patient population. Larotrectinib improved survival and quality adjusted survival vs entrectinib. Additional studies are needed to confirm these findings.