Clinical and Genetic Implications of DNA Mismatch Repair Deficiency in Patients With Pancreatic Ductal Adenocarcinoma

Abstract

This cohort study examines the behavior and outcomes of deficient DNA mismatch repair in patients at elevated risk for malignant neoplasms, including pancreatic ductal adenocarcinoma.

DNA mismatch repair status is a well-established biomarker in colorectal cancer and is associated with both a favorable prognosis and excellent response to immunotherapy. However, the influence of deficient DNA mismatch repair (dMMR) and microsatellite instability in pancreatic ductal adenocarcinoma (PDAC) remains unknown. The prognosis of PDAC is typically dismal and treatment options are limited, but previous research has suggested that microsatellite instability in resected PDAC specimens may be associated with improved survival. Because patients with Lynch syndrome harbor germline defects in mismatch repair genes and have significantly elevated lifetime risks of malignant neoplasms, including PDAC, we used this unique patient cohort to establish the clinical behavior and outcomes of dMMR PDAC.

Methods

The prospectively maintained Familial High-Risk Gastrointestinal Cancer Clinic and Genetic Counseling databases were queried to identify all patients with PDAC who were evaluated at the University of Texas MD Anderson Cancer Center from January 1, 2006, to December 31, 2016. Patients whose tumors showed evidence of dMMR by results of immunohistochemical analysis and/or who carry germline mutations in MLH1 (GenBank NG_007109), MSH2 (GenBank NG_007110), MSH6 (GenBank NG_007111), or PMS2 (GenBank NG_008466) were included. Clinicopathologic and genetic features as well as long-term follow-up were reviewed.

The MD Anderson Cancer Center Institutional Review Board approved this retrospective study and waived the individual informed consent requirement.

Results

Among 821 patients included in the registry, 10 (1.2%) patients with PDAC were identified (Table). Seven patients had pathogenic germline mutations in MLH1 (n = 4) and MSH2 (n = 3). The mean (SD) age of all patients was 57.2 (15.0) years, 7 patients were men, and 8 had a personal history of Lynch syndrome–associated cancers. Six patients presented with localized disease and underwent either pancreatoduodenectomy (n = 4) or distal pancreatectomy (n = 2) with curative intent. Among these 6 patients, adjuvant or neoadjuvant chemotherapy was delivered to 3 and chemoradiotherapy was provided to 2. The 5-year overall survival rate of these patients was 100% after a median follow-up of 93.1 months (95% CI, 52.5-201.7). Four patients who presented with metastatic disease and were treated with systemic chemotherapy alone had a median overall survival of 16.5 months (95% CI, 7.2-24.0) and a 5-year overall survival rate of 25% (Figure).

Table. Complete Clinical, Genetic, and Outcomes Information for Patients With Mismatch Repair Deficient Pancreatic Ductal Adenocarcinoma.

Age, y/Sex	Personal History of Other Malignant Neoplasms	PDAC Staging	Treatment	Outcome	Tumor IHC Loss of Expression	MSI	Germline Testing Results
70s/M	Sebaceous adenoma	pT3N1M0	Distal pancreatectomy, adjuvant gemcitabine hydrochloride/capecitabine, vaccine trial	Retroperitoneal recurrence (26 mo); 113.4 mo alivea	MSH2, MSH6	NA	MSH2 A636P (1906G>C)
70s/M	Rectal cancer (twice), colon cancer, sebaceous adenoma, small bowel lymphoma	ypT3N0M0	Neoadjuvant chemoradiotherapy (gemcitabine), Whipple procedure	Died of other causes; 72.8 mo survival	MLH1, PMS2b	NA	MLH1 (c.1731G>A)
40s/M	Sebaceous adenoma, colon cancer (twice)	Unknown	Whipple procedure	No evidence of disease; 334.6 mo alive	MLH1, PMS2b	Highb	MLH1 (deletion exon 16)
20s/F	None	TxNxM0	Whipple procedure	Retroperitoneal recurrence (69 mo)c; 201.8 mo survival	Not tested	NA	MLH1 (R265C)
40s/M	Colon cancer	pT1N0M0	Distal pancreatectomy, adjuvant gemcitabine	No evidence of disease; 18.6 mo alive	MLH1, PMS2	NA	MLH1 (deletion exon 1-3)
50s/F	Endometrial cancer	TxNxM1	Gemcitabine/capecitabine	Died of disease; 15.6 mo survival	MSH2b	NA	MSH2 R117X (c.2131C>T)
50s/M	Sebaceous adenoma, duodenal cancer	TxNxM1	Gemcitabine-abraxane	Died of disease; 8.9 mo survival	MSH2, MSH6b	Highb	MSH2 (c.2005 + 2dupT)
50s/F	Breast cancer, ovarian cancer	TxNxM1	Gemcitabine-cisplatin	Died of disease; 69.4 mo survival	MSH2, MSH6	NA	Uninformative negative
70s/M	Colon cancer	ypT3N0M0	Neoadjuvant FOLFIRINOX and chemoradiotherapy (gemcitabine), Whipple procedure	No evidence of disease; 52.5 mo alive	MSH2, MSH6b	Highb	Uninformative negative
60s/M	None	T4N1M1	FOLFIRINOX, gemcitabine-abraxane	Died of disease; 17.5 mo survival	MLH1, PMS2	NA	Uninformative negative

Abbreviations: FOLFIRINOX, fluorouracil, leucovorin calcium, irinotecan hydrochloride, and oxaliplatin; IHC, immunohistochemistry; MSI, microsatellite instability; NA, not applicable.

^aSlowly progressive retroperitoneal lymphadenopathy; initially observed and then treated with chemotherapy (gemcitabine/capecitabine) followed by chemoradiotherapy (50 Gy in 25 fractions with concurrent capecitabine).

^bTested on prior Lynch syndrome–associated tumor.

^cRetroperitoneal lymphadenopathy treated with chemotherapy (gemcitabine, oxaliplatin, bevacizumab) followed by chemoradiotherapy (50.4 Gy in 28 fractions with capecitabine).

Figure. Long-term Outcomes of Patients With Mismatch Repair Deficient Pancreatic Ductal Adenocarcinoma.

Six patients had localized disease and 4 patients had metastatic disease.

Discussion

In this single-institution study of dMMR PDAC, we identified patients with defects in MLH1, MSH2, MSH6, and PMS2 genes due to both germline and sporadic causes. Although patients with dMMR were not immune to nodal and distant metastases, those with localized disease who underwent surgery with curative intent had remarkably good outcomes with a 5-year overall survival rate of 100%, which is significantly better than the 25% reported in the largest randomized clinical trial of surgical resection and adjuvant chemotherapy. Patients who presented with metastatic disease also experienced better-than-expected outcomes, achieving a median of 16.5 months’ survival with one 5-year survivor (95% CI, 7.2-24.0) compared with a median of 11.1 months in patients treated with FOLFIRINOX (fluorouracil, leucovorin calcium, irinotecan hydrochloride, and oxaliplatin) and 6.8 months in patients treated with gemcitabine hydrochloride. These findings confirm those of previous studies, which suggested superior outcomes in patients with dMMR malignant neoplasms, presumably related to a more robust tumor-infiltrating immune response.

The limitation of this study is its retrospective, single-institution design and the inclusion of only patients referred to and evaluated in our familial and high-risk cancer clinic, but these findings have several implications. First, the diagnosis of PDAC in patients with a personal history of Lynch syndrome–associated cancers should prompt dedicated tumor and germline genetic testing. Second, treatment decisions should be made in the context of these patients’ favorable tumor biological characteristics and life expectancy. In fact, 2 patients in this series underwent local therapy (surgery or chemoradiotherapy) for retroperitoneal recurrences and experienced durable responses. Finally, previous research in colorectal cancer has demonstrated the influence of dMMR status on sensitivity to specific treatments, including checkpoint inhibitors. Ongoing clinical trials in patients with PDAC and mutations in DNA damage-repair genes, such as BRCA, PALB2, ATM, and MMR, hold promise for more effective targeted treatment options. Given the clinical significance of the findings reported herein, routine screening of PDAC specimens for dMMR may be indicated and should be the focus of future investigations.