• 2018-07
  • 2018-10
  • 2018-11
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2019-12
  • 2020-01
  • 2020-02
  • 2020-03
  • 2020-04
  • 2020-05
  • 2020-06
  • 2020-07
  • 2020-08
  • 2020-09
  • 2020-10
  • 2020-11
  • 2020-12
  • 2021-01
  • 2021-02
  • 2021-03
  • 2021-04
  • 2021-05
  • 2021-06
  • 2021-07
  • 2021-08
  • 2021-09
  • 2021-10
  • 2021-11
  • 2021-12
  • 2022-01
  • 2022-02
  • 2022-03
  • 2022-04
  • 2022-05
  • 2022-06
  • 2022-07
  • 2022-08
  • STF 31 We reviewed the results of comprehensive genomic prof


    We reviewed the results of comprehensive genomic profiling (CGP) in 26,054 cases of NSCLC and examined the frequency and clinical characteristics of patients whose tumors harbor FGFR fusions. To our knowledge, this is the most in-depth evaluation of FGFR fusions exclusively in NSCLC. Here we will describe 52 NSCLC cases harboring 14 unique FGFR fusion pairs.
    Methods The Foundation Medicine CGP platform was used in this analysis, with the second version of the assay updating the number of abnormalities evaluated per sample. In brief, at least 50 ng of DNA was extracted from formalin-fixed paraffin-embedded sections and CGP was performed on hybridization-captured, adaptor ligation–based libraries to a mean coverage depth of greater than 600× for 235 (version 1) or 315 (version 2) cancer-related genes (including FGFR1, FGFR2, FGFR3, and FGFR4) plus select introns from 19 (version 1) or 28 (version 2) genes that are frequently rearranged in cancer, as previously described. A minimum median coverage of more than 150× was required for each sample, and at least 99% of baits (intron and exon) were required to have more than 100× coverage. Sequencing coverage for FGFR1, FGFR2, FGFR3, and FGFR4 is detailed in Table 1. Cases positive for FGFR1, FGFR2, FGFR3, and FGFR4 fusions retaining the kinase domain were identified. Additional FGFR rearrangements predicted to disrupt the kinase domain or rearrangements without identifiable fusion partners were excluded from this analysis. In all, samples from 26,054 consecutive NSCLC cases were submitted as part of routine care to Foundation Medicine (Cambridge, MA), which is a Clinical Laboratory Improvement Amendments–certified and College of American Pathologists–accredited reference laboratory, from May 2012 to September 2017. In this study, 3531 samples were evaluated with version 1 and 22,523 samples were evaluated with version 2. As part of the assay, tumor mutational burden (TMB) was also determined for each sample. TMB was calculated as the total number of relevant mutations divided by the coding region target territory of the test (0.83 megabases [Mb] for version 1 and 1.14 Mb for version 2) and is characterized as the number of somatic STF 31 substitutions or indel alterations per Mb after filtering to remove known somatic and deleterious mutations.21, 22 Clinical data such as age, sex, histologic type, and stage were abstracted from reports accompanying each formalin-fixed, paraffin-embedded specimen. Ordinal relationships were examined by using the Mann-Whitney U test; categorical relationships were examined by using the Pearson chi-square test with Yates' continuity correction applied when applicable. Approval for this study, including a waiver of informed consent and a Health Insurance Portability and Accountability Act waiver of authorization, was obtained from the Western Institutional Review Board (protocol No. 20152817).
    Discussion To the best of our knowledge, our cohort of 26,054 patients with NSCLC is the largest to date to be interrogated specifically for FGFR fusions, which were identified in 0.2% of samples. FGFR fusions were seen more frequently in SCC (0.59%) than in adenocarcinoma (0.12%). Although the incidence in our cohort was lower than that reported in another smaller study (1% for all NSCLC), our results are consistent with reports that that FGFR fusions occur more frequently in lung tumors with squamous cell histologic features. This is of particular interest, as the discovery of targetable driver mutations with therapeutic implications has been largely limited to adenocarcinoma. In this study, we limited analysis to FGFR fusions retaining the kinase domain of FGFR fused to an identifiable fusion partner. Consistent with prior reports, fusions with FGFR as the 5' partner (n = 48) were more common than those with FGFR as the 3' partner (n = 4). Of note, additional cases in our NSCLC data set were observed to have rearrangements or truncations in FGFR intron 17 or exon 18 without an identifiable fusion partner, and these alterations may also be activating. Further, only select introns of the FGFR genes were sequenced, as noted in Table 1; therefore, it is possible that additional fusions with noncanonical breakpoints may be underrepresented. In this study 14% of cases were sequenced by using a version 1 of the CGP platform, which did not specifically include intron baiting for FGFRs (although approximately half of intron 17 was effectively baited for FGFR1, FGFR3, and FGFR4 because of overhang from exon baiting). In these samples, FGFR fusions were detected in five of 3531 samples (0.14%) versus in 47 of 22,523 samples (0.21%) tested by using version 2 of the assay, which includes some intron baiting of all four FGFRs including hotspot fusion breakpoints (Table 1). The overall difference in FGFR fusion frequency between versions 1 and 2 was not statistically significant (p = 0.53), but FGFR2 fusions in particular may have be underrepresented in samples tested by using version 1.