Fluorescent In Situ Hybridization Must be Preferred to pan-TRK Immunohistochemistry to Diagnose NTRK3-rearranged Gastrointestinal Stromal Tumors (GIST)

Tyrosine kinase inhibitors have revolutionized the treatment of patients with gastrointestinal stromal tumors (GISTs). Nevertheless, some GISTs do not contain any targetable KIT or PDGFRA mutations classically encountered in this field. Novel approved therapies targeting TRK chimeric proteins products of NTRK genes fusions consist in a promising approach to treat some patients with GISTs lacking any identified driver oncogenic mutation in KIT, PDGFRA or BRAF genes. Thus, an adequate testing strategy permitting to diagnose the rare NTRK-rearranged GISTs is required. In this work, we studied about the performances of pan-TRK immunohistochemistry (IHC) and NTRK1/2/3 fluorescent in situ hybridization in a series of 39 GISTs samples. Among 22 patients with GISTs lacking KIT or PDGFRA mutations, BRAFV600E IHC permitted to diagnose 2/22 (9%) BRAFV600E-mutated GISTs and, among the 20 KIT, PDGFRA, and BRAF wild type tumors, 1/20 (5%), NTRK3-rearranged tumor was diagnosed using NTRK3 fluorescent in situ hybridization. Pan-TRK IHC using EPR17341 and A7H6R clones was negative in this NTRK3-rearranged sample. Pan-TRK IHC was frequently positive in NTRK not rearranged tumors without (24 samples analyzed) or with (15 samples analyzed) KIT or PDGFRA mutations with major discrepancies between the 2 IHC clones (intraclass correlation coefficient of 0.3042). Given the new therapeutic opportunity offered by anti-TRK targeted therapies to treat patients with advanced cancers including GISTs, it is worth to extend molecular analysis to NTRK fusions testing in KIT, PDGFRA, and BRAF wild type GISTs. Pan-TRK IHC appears not relevant in this field but performing a simple NTRK3 fluorescent in situ hybridization test consists in a valuable approach to identify the rare NTRK3-rearranged GISTs treatable using anti-TRK therapies.

Use of protease inhibitors among persons with AIDS in Los Angeles County

We evaluated the access to, and the factors associated with, protease inhibitor use among persons with AIDS in Los Angeles County. A population-based sample of adult persons with AIDS is routinely interviewed in Los Angeles County and includes a 30% random sample of men who have sex with men and all other persons reported with AIDS. Since May of 1996, all participants were asked if their physician had ever prescribed a protease inhibitor for their use. The possible association between protease inhibitor use and sociodemographic, temporal and health care factors was assessed for the 12-month period May 1996 through April 1997. Logistic regression was employed for multivariate analysis. Over the 12-month study period, 61.7% (209/339) persons interviewed reported that their physician had prescribed a protease inhibitor as part of their therapy. In bivariate analysis, treatment with protease inhibitor use was more common for whites (71.4%) and US-born Latinos (68.2%) than blacks (53.4%) and foreign-born Latinos (56.6%), among person of higher income (71.2%) than lower (< $10,000) income (50.3%), in those who reported having insurance (66.7%) than those uninsured (47%) and among persons receiving care at private clinics (86.4%) than at HMOs (63.4%) or public clinics (55.2%). An increasing trend of protease inhibitor use with higher educational level and declining CD4+ count was observed. A temporal increase was noted and this trend was most pronounced for persons receiving care at public clinics. In multivariate analysis, persons receiving care at private facilities (adjusted OR = 2.9, 95% CI 1.0, 8.2) and those with higher incomes (adjusted OR = 2.5, 95% CI 1.5, 4.3), were more likely to report that their physician had prescribed a protease inhibitor. The effect of facility type was modified by time. During the first six months of the study period (May 1996-October 1996) persons with AIDS receiving care at public facilities and HMO sites were substantially less likely to report having been offered a protease inhibitor (adjusted OR = 0.13, 95% CI 0.03, 0.58 and adjusted OR = 0.23, 95% CI 0.05, 1.2, respectively). However, no significant facility-specific differences were observed over the last six-month period (November 1997-April 1997) evaluated. Our findings suggest that substantial differences exist in the prescribing and use of protease inhibitors among persons with AIDS in Los Angeles County. Several factors, including facility of HIV care, calendar time, income, education level and level of immunosuppression were independently associated with protease inhibitor use and suggest the existence of important barriers to access. Efforts should be made to identify and remove barriers that will ensure the widest possible access to protease inhibitors for patients with a clinical indication for their use.