Table 2.

Clinical efficacy according to F1L CDx assay and CTA results. IQR, interquartile range.

	CTA+ a	CTA+ F1L CDx evaluable				CTA+ F1L CDx unevaluable	P‐values d
		CTA+ F1L CDx+	CTA+ F1L CDx−	P‐values c	Total
Patients with an NTRK‐fp solid tumour
ORR (95% CI)	57.4 (43.2–70.8) n = 54	72.2 (46.5–90.3) n = 18	55.0 (31.5–76.9) n = 20	0.446	63.2 (47.3–76.6) n = 38	43.8 (19.8–70.1) n = 16	0.24
Median DoR, months (IQR)	10.4 (5.7–15.1) n = 54	9.2 (5.8–12.8) n = 18	12.9 (6.7–14.0) n = 20	0.434	9.3 (5.7–12.9) n = 38	14.1 (8.9–19.2) n = 16	0.40
Patients with ROS1‐fp NSCLC
ORR (95% CI)	78.4 (64.8–88.7) n = 51 b	72.2 (46.5–90.3) n = 18	72.7 (39.0–94.0) n = 11	1.00	72.4 (54.3–85.3) n = 29	86.4 (65.1–97.1) n = 22	0.31
Median DoR, months (IQR)	12.0 (5.6–17.2) n = 51	5.6 (3.5–11.4) n = 18	17.3 (13.9–18.8) n = 11	0.009	10.4 (3.7–17.2) n = 29	13.3 (8.5–16.4) n = 22)	0.31

^a ORR values in the CTA⁺ groups were derived in the ALKA‐372‐001/STARTRK‐1/STARTRK‐2 integrated analysis (May 2018 cut‐off) [17, 20].

^b Two ROS1‐fp patients were removed per FDA request.

^c P‐values derived with Wilcoxon rank‐sum test, for comparison between F1L CDx⁺ and F1L CDx⁻ groups.

^d P‐values derived from Fisher exact test for categorical factors between the F1L CDx evaluable set and the F1L CDx unevaluable set.