RESUMO
Objective: The vast majority of gastrointestinal stromal tumors (GISTs) are driven by activating mutations in KIT, PDGFRA, or components of the succinate dehydrogenase (SDH) complex (SDHA, SDHB, SDHC, and SDHD genes). A small fraction of GISTs lack alterations in KIT, PDGFRA, and SDH. We aimed to further characterize the clinical and genomic characteristics of these so-called "triple-negative" GISTs. Methods: We extracted clinical and genomic data from patients seen at MD Anderson Cancer Center with a diagnosis of GIST and available clinical next generation sequencing data to identify "triple-negative" patients. Results: Of the 20 patients identified, 11 (55.0%) had gastric, 8 (40.0%) had small intestinal, and 1 (5.0%) had rectal primary sites. In total, 18 patients (90.0%) eventually developed recurrent or metastatic disease, and 8 of these presented with de novo metastatic disease. For the 13 patients with evaluable response to imatinib (e.g., neoadjuvant treatment or for recurrent/metastatic disease), the median PFS with imatinib was 4.4 months (range 0.5-191.8 months). Outcomes varied widely, as some patients rapidly developed progressive disease while others had more indolent disease. Regarding potential genomic drivers, four patients were found to have alterations in the RAS/RAF/MAPK pathway: two with a BRAF V600E mutation and two with NF1 loss-of-function (LOF) mutations (one deletion and one splice site mutation). In addition, we identified two with TP53 LOF mutations, one with NTRK3 fusion (ETV6-NTRK3), one with PTEN deletion, one with FGFR1 gain-of-function (GOF) mutation (K654E), one with CHEK2 LOF mutation (T367fs*), one with Aurora kinase A fusion (AURKA-CSTF1), and one with FANCA deletion. Patients had better responses with molecularly targeted therapies than with imatinib. Conclusions: Triple-negative GISTs comprise a diverse cohort with different driver mutations. Compared to KIT/PDGFRA-mutant GIST, limited benefit was observed with imatinib in triple-negative GIST. In depth molecular profiling can be helpful in identifying driver mutations and guiding therapy.
RESUMO
Distinguishing well-differentiated liposarcoma (WDLPS) from dedifferentiated liposarcoma (DDLPS) is essential given distinct treatment paradigms and chemosensitivity. Percutaneous biopsy has a low sensitivity for detecting DDLPS. We sought to identify the diagnostic utility of positron emission tomography/computed tomography (PET/CT) in identifying WDLPS versus DDLPS. An independent radiologist reviewed PET/CT images to identify target lesions and determine the maximum standardized uptake value (SUVmax). An independent pathologist review confirmed WDLPS or DDLPS histology. A binary cutoff point of SUVmax was identified using a classification and regression trees (CART) algorithm. We identified 20 patients with WDLPS or DDLPS with 26 PET/CTs performed for separate recurrences that were followed by surgical sampling. Of the 26 records, 12 were DDLPS (46%) and 14 were WDLPS (54%). Patients with DDLPS had significantly higher SUVmax than those with WDLPS (p value = 0.0035). A SUVmax of 4 was identified as the cutoff point. Using this cutoff, the sensitivity of SUVmax identifying a case as DDLPS was 83.3% (95% CI: 51.6%, 97.9%) and the specificity was 85.7% (95% CI: 57.2%, 98.2%). PET/CT is a sensitive and specific diagnostic tool to identify the presence of dedifferentiation within the tumor.
