Abstract
Purpose
Uncertainty about the prognostic significance of neurotrophic tropomyosin receptor kinase (
NTRK
) fusions complicates clinical decisions and interpretation of survival outcomes in oncology care. Drawing on the largest cohort of
NTRK
fusion–positive (
NTRK
+) patients, we establish the prognostic effect of
NTRK
fusions in adults with advanced cancers.
Methods
We used a large real-world clinicogenomic research database from the United States to emulate a pragmatic randomized trial that determines the causal effect of harboring an
NTRK
gene fusion (
NTRK
+) on survival prognosis for adult patients with advanced solid tumors. We 1:1 matched
NTRK
+ patients to patients with no identified
NTRK
genomic alterations (
NTRK
–) using a machine learning matching approach. Excluding patients who received TRK inhibitors, we conducted stratified matching within common tumor groups. We used Kaplan-Meier and Weibull regression survival analyses and estimated the difference in restricted mean survival time (RMST) over 1-, 2-, 3-, and 5-year study periods.
Results
We included 68,145
NTRK–
patients for matching to 73
NTRK
+ patients within seven tumor groups, including colorectal (n = 22), non–small cell lung (n = 13), sarcoma (n = 13), breast (n = 10), head and neck (n = 7), neuroendocrine (n = 4), and prostate (n = 4), and 14
NTRK
+ patients representing eight other cancers. The difference in overall survival over 1 year was statistically significant using the Kaplan-Meier estimate (
P
= .008), Weibull regression (hazard ratio, 2.105 [95% CI, 1.207 to 3.673]), and RMST (–50 days [95% CI, –96 to –5];
P
= .029). Differences were also statistically different at 2 years and 3 years. At 5 years, no survival differences were statistically significant.
Conclusion
In this large real-world target trial emulation,
NTRK
gene fusions were associated with clinically meaningful reductions in overall survival among adults with advanced cancers not treated with TRK inhibitors. These findings clarify the prognostic role of
NTRK
fusions and inform clinical interpretation of outcomes in routine oncology.