br FISH assay was performed using
FISH assay was performed using the Zytolight SPEC ROS1 Dual Color Break Apart Probe (ZytoVision, Germany). This break-apart FISH test is based on a mixture of two probes hybridizing to the proximal (3’, green-labeled probe) and distal (5’, orange-labeled probe) to the ROS1 breakpoint cluster region. At least 50 non-overlapping tumor nuclei were scored for each specimen by a trained technologist and a pa-thologist. Cells positive for rearrangement were defined by two main patterns: i) a “split pattern”, with 3’ and 5’ break apart signals at a distance of two times the diameter of the largest signal; ii) a “5’ deletion pattern”, showing one fusion signal and an isolated 3’green signal (without the corresponding 5’ orange signal). A case was considered FISH positive for ROS1 rearrangements when at least 15% of tumor cells showed any split or any 5’ deletion pattern.
NGS was performed starting from 10 μm-thick serial sections or cytological smears including the tumor area of interest with at least 50% tumor cell enrichment. Slides were manually macrodissected, and the RNA was isolated. Library were prepared using the Oncomine™ Focus Assay, 318 Solution (Thermo Fisher Scientific) using a total of 10 ng input RNA per sample. The RNA panel can identify rearrange-ments in 23 genes including ROS1. Sequencing was performed using the Ion PGM™ Hi-Q™ View Sequencing Kit on the Ion Torrent Personal Genome Machine (Thermo Fisher Scientific). Fusions were detected using the fusion detection module within the Ion Reporter workflow, in particular 20,000 was the minimum number of total valid mapped reads required to qualify a sample as valid and to proceed with the analysis.
The mean age of the 8 patients was 56,5 years (range 46–67), 5 were males and 3 females. At diagnosis, none of the patients presented with Y-27632 metastasis. 2 were in stage IIIB, 5 in stage IVA for bilateral lung nodules and pleural eﬀusion, and one in stage IVB with multiple liver lesions. 5 patients were never-smokers, two light former smokers, one patient (case # 1) current heavy smoker (50 pack/year).
Both FISH and NGS analyses were successfully carried out in all patients. Table 1 summarizes the clinical and the laboratory findings in the eight patients. In particular, the break-apart FISH test revealed a 5’ ROS1 deletion in all the 8 cases, characterized by the presence of iso-lated green signals indicating the 3’probe (Fig. 1). The percentage of rearranged nuclei ranged from 30 to 60%. Concomitant “split” signals were encountered sporadically and never exceeded the 15% of cancer nuclei. In 4 of 8 cases (cases # 2,3,7,8) the NGS analysis confirmed a ROS1 fusion; in three cases with the partner EZR and in one with SDC4 Lung Cancer 135 (2019) 88–91
Still on treatment NO YES YES YES YES NO YES NO
Status D = dead A = Alive DAAAADAA
Response to crizotinib NO (PD) YES (CR) YES (CR) YES (PR) YES (PR) NO (PD) YES (PR) NO (PD)
Line of crizotinib administration Second First Second Second Second Second Second Second
ROS1 partner gene
of gene fusion
Presence NO YES YES NO NO NO YES YES
Fig. 1. NGS and FISH results and response to therapy related to case #8: A) NGS analysis report describing the fusion of ROS1 with SDC4 (Ion Reporter software and IGV visualization); B) FISH image showing the 5’ ROS1 deletion pattern (DAPI 100X); C) CT and (D) PET scan before (first column) and after (second column) therapy with Crizotinib.
Therapy with Crizotinib was started in all patients and lasted for a mean of 11.0 months (range 2–31). The median overall survival was not reached at a median follow up of 11.1 months (15.7 months censored only). Objective responses were observed in 5 of 8 patients, two with complete response, while the remaining three faced rapid disease pro-gression (Fig. 1). All the patients with confirmed ROS1/EZR re-arrangement at NGS displayed an objective response to Crizotinib and are still alive at the time of last available follow-up. The three patients who experienced rapid progressive diseases were either negative for ROS1 fusions (#1 and 6) or displayed a fusion with a gene diﬀerent from EZR at NGS (#8). Two patients (#4 and 5) showed an objective response to Crizotinib despite absent detection of ROS1 fusions at NGS but one of them (#5) harbored a concomitant ALK fusion.