Mutational Landscape in Resected Periampullary Adenocarcinoma: Relationship With Morphology and Clinical Outcome.
retrospective_cohort · Level III
Where this comes from
- Record sourced from PubMed, PMID 32914025.
- Also identified by DOI 10.1200/PO.18.00323 and PMC identifier 7446458.
- Licence recorded as CC BY.
- The licence permits redistribution, so the abstract is shown in full and the full text is available from the publisher.
Abstract
Periampullary adenocarcinomas encompass a heterogeneous group of tumors with dismal prognosis and limited treatment options. Emerging evidence shows that tumor morphology (ie, intestinal type [I-type] or pancreatobiliary type [PB-type]) is a more relevant prognostic factor than tumor origin. Knowledge is sparse, however, on whether key mutations differ according to morphology. Next-generation sequencing was applied to assess the mutational status of 70 genes in 102 tumors from a retrospective cohort of 175 patients with resected periampullary adenocarcinoma. Brahma-related gene 1 protein expression was examined by immunohistochemistry on tissue microarrays with primary tumors from the original cohort. <i>APC</i> mutations were significantly more common in I-type than in PB-type tumors (27.5% <i>v</i> 0%; <i>P</i> < .001), as were <i>ERBB3</i> mutations (20.8% <i>v</i> 4.8%; <i>P</i> = .016), whereas <i>CDKN2A</i> mutations were more common in PB-type than in I-type tumors (19.4% <i>v</i> 2.5%; <i>P</i> = .013). <i>KRAS</i> mutation was an independent factor of poor prognosis in I-type tumors (hazard ratio, 3.73; 95% CI, 1.10 to 12.67). In PB-type tumors, <i>SMARCA4</i> mutation was an adverse prognostic factor in patients not receiving adjuvant chemotherapy, and there was a significant treatment interaction between expression of Brahma-related gene 1 protein, the protein encoded by <i>SMARCA4</i>, and adjuvant chemotherapy (<i>P</i> <sub>interaction</sub> = .007). To our knowledge, this is the first description of the mutational landscape in the full spectrum of periampullary adenocarcinoma that demonstrates that the distribution and prognostic and predictive significance of commonly mutated genes differ by morphology. The results emphasize that morphology is an important factor to consider in the search for novel biomarkers and targeted personalized treatment of these patients. In addition, the findings support the concept that molecular profiling of these tumors could be of clinical benefit.