Testing For dMMR Research Articles

Overview of acomparative analysis of microsatellite instability and standard mismatch repair protein-deficiency tests in alarge cancer cohort.

Mismatch repair deficiency (dMMR) with microsatellite instability (MSI) is frequent in cancer, particularly in gastrointestinal and endometrial malignancies. The increased tumor mutational burden renders dMMR/MSI tumors suitable targets for immune checkpoint inhibitors-provided the regulatory genetic defect can be detected. dMMR and MSI are considered equally effective predictors of the efficacy of ICIs; however, while dMMR testing is based on detection of missing MMR proteins in immunohistochemistry (IHC), MSI polymerase chain reaction (PCR) testing focuses on the consequences of dMMR at the genomic level. Aretrospective analysis was carried out in alarge cancer cohort (n = 1306). dMMR was tested by four IHC reactions (MLH1, PMS2, MSH2, MSH6), and MSI was assessed by pentaplex PCR (BAT-25, BAT-26, MONO-27, NR-21, NR-24) in 703cases. In 64cases (5%), technical failures (mostly poor preanalytical fixation) prevented dMMR/MSI testing. Tumors were colorectal (CRC; n: 978), cancer of unknown primary (n: 126), endometrial (n: 39), pancreatic (n: 36), and gastric (n: 33). dMMR was diagnosed as classical, nonclassical, or unusual, depending on IHC. The MSI-high incidence was 12.1% overall and similar in the CRC subcohort. Interestingly, the dMMR incidence was higher in the total cohort (20.3%) and similar in the CRC subcohort. The incidences of proficient MMR (pMMR) and microsatellite stability (MSS) were similar in the total cohort and in the CRC subcohort. A19.3% discrepancy was found between MMR IHC and MSI PCR for the entire cohort, independent of tumor types. In the case of pMMR, the discrepancy rate for MSS/MSI-low was low (2.0%; entire cohort and the CRC subcohort). However, the discrepancy between dMMR and MSI-high was high within the entire cohort (60.9%) and in the CRC (58.6%) and non-CRC subcohorts (68%). This high discrepancy was not due to tumors with alow T/N ratio. Regarding dMMR phenotypes, classical dMMR had a~ 60% correlation with MSI-high status, while non-classical dMMR had amuch lower and unusual dMMR avery low (< 10%) correlation with MSI-high in the entire cohort and in the CRC subcohort. Overall, the MSI PCR sensitivity for MMR IHC status was very low. dMMR and MSI-high likely result in an increased rate of structurally altered proteins, i.e., neoantigens, and the efficacy of cancer immunotherapies is thus expected to be higher. We compared MMR IHC to MSI PCR in alarge cohort of cancer patients to study how PCR test results correlate to MMR IHC. Our data imply that preanalytical factors strongly influence the results of MMR IHC and MSI PCR and may question the current dogma that dMMR phenotype and genetic MSI status are equivalent predictive markers for immunotherapies.

Genomic Alterations in DNA Mismatch Repair Genes Across Different Cancer Types.

PD-1 inhibition is effective in patients with mismatch repair deficient (dMMR) solid tumors in a tumor-agnostic fashion. However, dMMR testing by immunohistochemistry (IHC) is not routinely performed across tumor types. By contrast, next-generation sequencing (NGS) for somatic genomic alterations is frequently performed across tumor types. We hypothesized that NGS would identify patients with alterations in mismatch repair (MMR) genes and that these patients would have higher rates of MMR protein loss by IHC. This would support the utility of IHC reflex testing after NGS and potential matching to approved therapeutic options. From January 2016 to December 2021, 15,701 patients with solid tumors received NGS covering the MMR genes, and 4,994 patients had both IHC and NGS. Sequencing results were analyzed for mutations in MMR genes, tumor type distribution, and concordance with IHC results when available. Six hundred and ninety-eight (4.4%) of 15,701 patients had mutations in one of the MMR genes. Mutations were found across tumor types. Three hundred and seventeen (6.3%) of 4,994 patients displayed IHC loss for at least one MMR protein. 33.8% patients (110/325) patients with MMR mutations had dMMR, compared with just 4.4% (207/4,669) patients without mutations (P < .001); dMMR rate varied by mutation type. Mutations in MMR genes are found in multiple tumor types where IHC testing is not routine. Reflex IHC testing of patients carrying MMR gene mutations, especially those known or inferred to be inactivating, may identify more patients with dMMR and matched treatment options. However, dedicated IHC screening is needed to capture majority of the patients.

Real world impact of pembrolizumab availability for deficient mismatch repair metastatic colorectal cancer.

e15569 Background: Keynote 177 was a practice changing study, with pembrolizumab versus dealers’ choice chemotherapy +/- biologic significantly improving progression free survival in patients (pts) with previously untreated deficient mismatch repair (dMMR) metastatic colorectal cancer (mCRC). A high rate (29.4%) of pembrolizumab treated pts had stopped treatment at or before first response evaluation due to progressive disease as best response. Pembrolizumab availability in routine care is anticipated to benefit individual pts otherwise treated with chemotherapy. Additional pts not fit for chemotherapy, but fit for first line (1L) pembrolizumab, may be able to receive active treatment, which could increase the dMMR mCRC population benefit. Pembrolizumab availability may also drive increased MMR testing, which may impact the proportion of pts with dMMR mCRC. Methods: We analysed data from the TRACC registry of pts with mCRC in Australia comparing patient treatment across two cohorts. The first cohort was pts diagnosed from 1/1/2015 to 31/7/2021 when immunotherapy was only available via trial, self-funding or through a limited access program and the second cohort was pts diagnosed from 1/8/2021 to 23/1/2024, following the government’s Pharmaceutical Benefits Scheme reimbursement (PBSr) of pembrolizumab as 1L treatment for dMMR mCRC. We also examined early treatment discontinuation rates for pembrolizumab treated pts after PBSr. Results: Of 2819 mCRC pts overall, 2344 (83%) had known MMR status. Of pts tested 163 (7%) were dMMR. Pts with dMMR mCRC were older (mean age 66.6 vs 62.2, p &lt; 0.01) and had more co-morbidities (59% vs 47% for modified Charlson Comorbidity Index ≥ 3, p &lt; 0.01). They were also more likely to have a right side primary (67% vs 31%, p &lt; 0.01), BRAF V600E mutation (49% vs 11%, p &lt; 0.01) and peritoneal or brain metastases (34% vs 22%, p &lt; 0.01). Before PBSr of pembrolizumab, 85 of 117 (73%) dMMR mCRC pts received 1L treatment; 62/85 (73%) chemotherapy, 19/85 (22%) immunotherapy and 4/85 (5%) combination BRAF inhibitor and EGFR inhibitor. Following PBSr, an increased proportion (40 of 46) of dMMR mCRC pts received any 1L treatment (87% vs 73%, p = 0.05), with 1/40 (2.5%) receiving chemotherapy and 39/40 (97.5%) pembrolizumab. In a landmark analysis of these 39 pts, 12 of 19 (63%) that had started treatment ≥ 6 months ago remained on pembrolizumab. Conclusions: A high proportion of Australian pts with mCRC are undergoing dMMR testing. Increasing testing, including of more elderly pts, may increase the proportion of mCRC pts that are dMMR compared to that reported for trial populations. dMMR mCRC pts have multiple adverse prognostic features. The broad availability of pembrolizumab, such as through the Australian PBSr, looks likely to lead to an increased proportion of pts receiving any 1L therapy. Early data suggests the expected clinical benefit, with a low rate of early treatment discontinuation.

Identification of people with Lynch syndrome from those presenting with colorectal cancer in England: baseline analysis of the diagnostic pathway

It is believed that >95% of people with Lynch syndrome (LS) remain undiagnosed. Within the National Health Service (NHS) in England, formal guidelines issued in 2017 state that all colorectal cancers (CRC) should be tested for DNA Mismatch Repair deficiency (dMMR). We used a comprehensive population-level national dataset to analyse implementation of the agreed diagnostic pathway at a baseline point 2 years post-publication of official guidelines. Using real-world data collected and curated by the National Cancer Registration and Analysis Service (NCRAS), we retrospectively followed up all people diagnosed with CRC in England in 2019. Nationwide laboratory diagnostic data incorporated somatic (tumour) testing for dMMR (via immunohistochemistry or microsatellite instability), somatic testing for MLH1 promoter methylation and BRAF status, and constitutional (germline) testing of MMR genes. Only 44% of CRCs were screened for dMMR; these figures varied over four-fold with respect to geography. Of those CRCs identified as dMMR, only 51% underwent subsequent diagnostic testing. Overall, only 1.3% of patients with colorectal cancer had a germline MMR genetic test performed; up to 37% of these tests occurred outside of NICE guidelines. The low rates of molecular diagnostic testing in CRC support the premise that Lynch syndrome is underdiagnosed, with significant attrition at all stages of the testing pathway. Applying our methodology to subsequent years’ data will allow ongoing monitoring and analysis of the impact of recent investment. If the diagnostic guidelines were fully implemented, we estimate that up to 700 additional people with LS could be identified each year.

Application of machine learning in the prediction of deficient mismatch repair in patients with colorectal cancer based on routine preoperative characterization.

Detecting deficient mismatch repair (dMMR) in patients with colorectal cancer is essential for clinical decision-making, including evaluation of prognosis, guidance of adjuvant chemotherapy and immunotherapy, and primary screening for Lynch syndrome. However, outside of tertiary care centers, existing detection methods are not widely disseminated and highly depend on the experienced pathologist. Therefore, it is of great clinical significance to develop a broadly accessible and low-cost tool for dMMR prediction, particularly prior to surgery. In this study, we developed a convenient and reliable model for predicting dMMR status in CRC patients on routine preoperative characterization utilizing multiple machine learning algorithms. This model will work as an automated screening tool for identifying patients suitable for mismatch repair testing and consequently for improving the detection rate of dMMR, while reducing unnecessary labor and cost in patients with proficient mismatch repair. Deficient mismatch repair (dMMR) indicates a sustained anti-tumor immune response and has a favorable prognosis in patients with colorectal cancer (CRC). Although all CRC patients are recommended to undergo dMMR testing after surgery, current diagnostic approaches are not available for all country hospitals and patients. Therefore, efficient and low-cost predictive models for dMMR, especially for preoperative evaluations, are warranted. A large scale of 5596 CRC patients who underwent surgical resection and mismatch repair testing were enrolled and randomly divided into training and validation cohorts. The clinical features exploited for predicting dMMR comprised the demographic characteristics, preoperative laboratory data, and tumor burden information. Machine learning (ML) methods involving eight basic algorithms, ensemble learning methods, and fusion algorithms were adopted with 10-fold cross-validation, and their performance was evaluated based on the area under the receiver operating characteristic curve (AUC) and calibration curves. The clinical net benefits were assessed using a decision curve analysis (DCA), and a nomogram was developed to facilitate model clinical practicality. All models achieved an AUC of nearly 0.80 in the validation cohort, with the stacking model exhibiting the best performance (AUC = 0.832). Logistical DCA revealed that the stacking model yielded more clinical net benefits than the conventional regression models. In the subgroup analysis, the stacking model also predicted dMMR regardless of the clinical stage. The nomogram showed a favorable consistence with the actual outcome in the calibration curve. With the aid of ML algorithms, we developed a novel and robust model for predicting dMMR in CRC patients with satisfactory discriminative performance and designed a user-friendly and convenient nomogram.

Impact of Institutional Universal Microsatellite-Instability (MSI) Reflex Testing on Molecular Profiling Differences Between Younger and Older Patients with Colorectal Cancer

IntroductionDNA mismatch repair deficient (dMMR) or microsatellite instability-high (MSI-H) colorectal cancer (CRC) is found in about 15% of early-stage diseases and 5% of metastatic diseases. We reviewed a large, single-institutional database after implementation of universal reflex dMMR/MSI-H testing in CRC to compare profiles of younger (≤50) and older (>50) patients. Patients and MethodsBetween 2009 and 2017, all patients diagnosed with CRC at the University of Florida underwent reflex somatic tumor testing for dMMR by immunohistochemistry (MLH1, PMS2, MSH2, MSH6), MSI by PCR, and Next-Generation Sequencing. Statistical analysis was conducted with 2-sample comparison tests and logistic regression models. ResultsThere were 375 patients included in the final analysis. Patients were grouped as younger (ages ≤50 years-old; n = 80) or older (>50 years-old; n = 295). Compared to tumors from older patients, tumors from younger patients were less likely to be dMMR/MSI-H (12.5% vs. 21.4%, P = .013) and less likely to have a BRAF mutation (1.5% vs. 16.1%, P = .002). BRAF mutation status was highly associated with MMR status; BRAF-mutated tumors were 29.7 times more likely than BRAF-WT tumors to be dMMR/MSI-H (P = < .001, 95% CI 11.3-78.3). ConclusionsTumors of younger patients were less likely than tumors of older patients to have a dMMR/MSI-H or BRAF mutation. Universal MMR/MSI testing in our dataset identified a relatively large population of older patients with sporadic CRC who were eligible for immunotherapy.

597P Endometrial carcinoma and mismatch repair deficiency: Clinical association and universal screening for Lynch syndrome

Endometrial carcinoma (EC) is characterised by a high rate of DNA Mismatch Repair deficiency (dMMR). Testing for dMMR might have implications for EC prognosis and is part of Universal Screening (US) for Lynch syndrome (LS) diagnosis. However, this is not widely implemented yet, and LS is still underdiagnosed. For these reasons, we aimed to evaluate dMMR prevalence, its association with clinicopathological variables and its role for LS diagnosis for EC patients. This is a monocentric retrospective study on 368 consecutive EC cases (stages I-IV). From 2013 to 2020 they underwent surgery and immunohistochemistry (IHC) analysis for dMMR. We collected age, date of surgery, tumor histotype, grading, FIGO stage at diagnosis, lymphovascular invasion (LVSI), and survival status. We then assessed the proportion of dMMR patients on the entire cohort. Major outcome were OS and RFS in the dMMR patients compared to MMR proficient cases (pMMR). Among the dMMR cases, we reported those who underwent genetic testing and had LS. dMMR was found in 91/368 EC patients (25%). 84 had endometroid histotype, 7 other histotypes (92% vs 8% p= 0.010), 50 had G2 nuclear grading vs 19 G1 and 22 G3 (55% vs 21% vs 24%; p=0.005). As compared to pMMR at a 4-year follow-up an unfavorable trend in OS and RFS was observed for patients with dMMR (4 y OS 82% vs 74% p=0.115, 4 y RFS 70% vs 54% p-value=0.153). In multivariate analysis, age (p<0.001), histotype and grade (p<0.001), stage (p=0.003) and LVSI (p=0.010) were independently associated to OS. dMMR only showed a trend in worse OS (p=0.10). Among the 91 patients with dMMR, only 5 underwent genetic counseling (GC)(6.6%): LS was diagnosed in 1 patient (20%) with loss of MSH2/MSH6 expression. This is the largest Italian cohort of US for LS. dMMR prevalence was 25%, mainly associated with endometrioid type, G2 grading and a worse 4 y OS, as compared to pMMR cases, consistent with the literature. Unfortunately, oncologists' compliance with US for LS is still very poor, with a 6.6% GC referral. This should be implemented according to international guidelines, to increase LS diagnosis, reduce mortality in relatives and incidence of second cancers in survived patients.

Mismatch repair deficiency in early-onset duodenal, ampullary, and pancreatic carcinomas is a strong indicator for a hereditary defect.

Mismatch repair deficiency (dMMR) is a hallmark of Lynch syndrome (LS), but its prevalence in early‐onset (diagnosed under the age of 50 years) duodenal, ampullary, and pancreatic carcinomas (DC, AC, and PC, respectively) is largely unknown. We explored the prevalence of dMMR and the underlying molecular mechanisms in a retrospectively collected cohort of 90 early‐onset carcinomas of duodenal, ampullary, and pancreatic origin. dMMR was most prevalent in early‐onset DCs (47.8%); more than half of those were associated with hereditary cancer syndromes (LS or constitutional mismatch repair deficiency syndrome). All dMMR AC and PC were due to LS. Concordance of dMMR with underlying hereditary condition warrants ubiquitous dMMR testing in all early‐onset DC, AC, and PC.

Gynecological Cancers Caused by Deficient Mismatch Repair and Microsatellite Instability.

Simple SummaryMicrosatellite instability (MSI) has been detected in multiple types of gynecologic cancers. MSI is linked to mutations in mismatch repair (MMR) genes that cause mismatch repair deficit (dMMR) in human cells. Discovery of new therapeutic approaches are needed especially for treatment of advanced endometrial and other gynecological cancers with dMMR/MSI. In addition, there is a need to identify markers for reliable detection of dMMR/MSI gynecological cancers. Determination of the mechanism leading to these malignancies would help in diagnosis and therapeutic intervention. In this review, we summarize the MMR defects and MSI observed in gynecological cancers, and new therapeutic strategies to treat these cancers.Mutations in mismatch repair genes leading to mismatch repair (MMR) deficiency (dMMR) and microsatellite instability (MSI) have been implicated in multiple types of gynecologic malignancies. Endometrial carcinoma represents the largest group, with approximately 30% of these cancers caused by dMMR/MSI. Thus, testing for dMMR is now routine for endometrial cancer. Somatic mutations leading to dMMR account for approximately 90% of these cancers. However, in 5–10% of cases, MMR protein deficiency is due to a germline mutation in the mismatch repair genes MLH1, MSH2, MSH6, PMS2, or EPCAM. These germline mutations, known as Lynch syndrome, are associated with an increased risk of both endometrial and ovarian cancer, in addition to colorectal, gastric, urinary tract, and brain malignancies. So far, gynecological cancers with dMMR/MSI are not well characterized and markers for detection of MSI in gynecological cancers are not well defined. In addition, currently advanced endometrial cancers have a poor prognosis and are treated without regard to MSI status. Elucidation of the mechanism causing dMMR/MSI gynecological cancers would aid in diagnosis and therapeutic intervention. Recently, a new immunotherapy was approved for the treatment of solid tumors with MSI that have recurred or progressed after failing traditional treatment strategies. In this review, we summarize the MMR defects and MSI observed in gynecological cancers, their prognostic value, and advances in therapeutic strategies to treat these cancers.

Impact of universal immunohistochemistry on Lynch syndrome diagnosis in an Australian colorectal cancer cohort

Current guidelines recommend a step-wise screening algorithm for all colorectal carcinomas (CRC) to identify patients with Lynch syndrome (LS). To describe the frequencies of mismatch repair deficiency (dMMR), BRAFV600E mutations and MLH1 methylation in resected CRC, and evaluate the impact of universal screening on LS detection. Retrospectively, 1171 consecutive cases of resected CRC were identified between 2010 and 2017 from a large multi-centre pathology service. Testing for dMMR by immunohistochemistry (IHC) was initiated by the reporting pathologist from 2010, until universal testing was introduced in 2015. Patients with dMMR were referred to the Family Cancer Clinic (FCC) for consideration of germline mutation analysis. IHC was performed on 680 tumours, with abnormal expression in 124 (18%). Referral to FCC was made for 44 of the 88 patients with abnormal IHC (excluding those with BRAFV600E mutations). Of the 29 who attended, 16 underwent germline genetic testing, and LS was diagnosed in 7 with a germline mutation. After implementation of universal testing, there was a greater incidence of dMMR (17% vs 10%, P = 0.02), rate of BRAFV600E testing (79% vs 25%, P < 0.0001), and referral to FCC (61% vs 33%, P < 0.0001), but no difference in FCC attendance rate (65% vs 67%, P = 0.59) or new LS diagnoses (1.6% vs 0%, P = 0.06). Universal IHC testing may increase the detection of LS, and should be implemented where possible. However, the full benefit was limited by low referral to and uptake of genetic testing, and further strategies are needed to overcome these barriers.

Japan Society of Clinical Oncology provisional clinical opinion for the diagnosis and use of immunotherapy in patients with deficient DNA mismatch repair tumors, cooperated by Japanese Society of Medical Oncology, First Edition

BackgroundNovel therapeutic agents have improved survival outcomes in patients with advanced solid tumors. In parallel, the development of predictive biomarkers to identify patients who are likely to benefit from a certain treatment has also contributed to the improvement of survival. Recently, clinical trials have reported the efficacy of immune checkpoint inhibitors in the treatment of mismatch repair-deficient (dMMR) advanced solid tumors. In Japan, a PD-1 inhibitor for dMMR advanced solid tumors, regardless of the primary tumor site, has been approved. However, there are some issues related to administering immune checkpoint inhibitors in the clinical practice setting, making it necessary to develop the guidelines.MethodsClinical questions (CQs) regarding medical care were formulated for patients with dMMR advanced solid tumors, and evidence to the CQs was collected by manual search to prepare recommendations. Then, the committee members voted to determine the level of each recommendation considering the strength of evidence, expected risks and benefits to patients, and other factors.ResultsThe current guideline, which we consider a provisional clinical opinion at this point, describes the 11 requirements to be considered in terms of patients for whom dMMR testing is recommended, the timing and methods of dMMR testing, and clinical care systems required to perform dMMR testing properly and to administer immune checkpoint inhibitors safely.ConclusionThis provisional clinical opinion proposes the requirements for performing dMMR testing properly to select patients who are likely to benefit from immune checkpoint inhibitors and administering them safely.

Immunotherapy with checkpoint inhibitors in mismatch repair deficient (dMMR) cancers: Looking for a needle in a haystack— But, joy on finding one.

e14073 Background: Pembrolizumab is approved for deficient mismatch repair (dMMR) or high microsatellite instability (MSI-H) metastatic solid tumors with objective response rate (ORR) of 39.6% and response lasting for more than 6 months in 78% patients. Nivolumab is also approved in dMMR metastatic colorectal cancers (mCRC) with ORR of 28% and response lasting greater than 6 months in 67% patients. But finding these patients is difficult as only about 5% of metastatic cancers have dMMR. There is limited data from India in this population. We report our experience with testing and treatment in these patients. Methods: This is a single centre, retrospective study of metastatic solid tumors which had progressed on standard treatment. 137 patients were tested between May 2017 and Dec 2018. MMR testing was done by immunohistochemistry (IHC). The aim was to identify patients with dMMR; and to see their ORR, progression free survival (PFS) at 1 year, and adverse events on treatment with Nivolumab. Results: 137 metastatic cancer patients were tested. 75 had colorectal cancer (mCRC), 30 gastric, 10 hepatobiliary, 10 pancreatic, 8 endometrial, 2 small bowel cancer and 2 had breast cancer. 15 (11%) patients had dMMR. 3 (37%) of endometrium, 5 (16.6%) of stomach, 6 (8%) of mCRC and 1(10%) of hepatobiliary tree cancer had dMMR. 5 of the 15 dMMR patients received immunotherapy with Nivolumab - 2 in mCRC, 2 in stomach and 1 in endometrium. ORR was 60% with 3 partial responses, stable disease was seen in 1 (20%) and progression in 1 (20%). PFS at 1 year was 80%. Treatment was well tolerated.1 patient had hypothyroidism and 1 patient had grade 1 skin toxicity. Conclusions: Testing for dMMR is important in metastatic solid tumors as these patients are ideal for treatment with immunotherapy. But finding dMMR is difficult due to its infrequent presentation, and has been seen in 11% of our unselected patients. We recommend testing for dMMR by IHC in our Indian patients, as this is fast and cost effective. Finding dMMR cancers, and then treating with immunotherapy is rewarding irrespective of the site of origin. High ORR of 60% and 1 year PFS of 80% is very heartening to see in this relapsed metastatic patient group. Treating more dMMR patients and longer followup, will further elucidate the benefit of immunotherapy in our patients.

Results from an institutional universal reflex testing program for MSI/MMR in colorectal cancer.

796 Background: In colorectal cancer (CRC), clinical guidelines and immunotherapy treatment selection requires knowledge of tumor DNA mismatch repair gene deficiencies (dMMR) or accumulation of microsatellite repeats through genomic errors (MSI-H). Tumors harboring dMMR/MSI-H are found in 15- 20% of early stage CRC while the prevalence is ~5% in metastatic disease. Reflex testing of CRC to identify these important subsets has been proposed as a system-solution to improve identification. We present a large, single-institutional database of CRC reflexively profiled for MSI/MMR status at the University of Florida (UF). Methods: Beginning in 2009, stage IV CRC underwent reflex testing for MSI/MMR status. Earlier stage CRC testing began in subsequent years. For all new CRC diagnosed at UF, concurrent testing for dMMR by IHC (MLH1, PMS2, MSH2, MSH6), MSI by PCR (Promega MSI kit) and NGS is performed with appropriate positive and negative controls. IHC protein loss is confirmed by second GI pathologist. MSI is not performed if inadequate adjacent normal tissue. We conducted a retrospective analysis of all CRC samples analyzed between 2009 and 2017. Clinical data was collected from the EMR. This study was approved by the UF IRB. Results: A total of 388 new CRC cases were reflex tested (16% Stage I, 23% Stage II, 35% Stage III, and 25% Stage IV). Median age at diagnosis was 63 yrs (range: 17-98 yrs), 51% male and 79% white. Both MMR and MSI were performed in 244 (63%) tumors with 100% concordance when concurrently tested. dMMR/MSI-H incidence (20% in overall population) decreased with stage: I/II (31%), III (18%) and IV (~9%) and was associated with BRAF V600E mut in 36/76 cases (47%). Importantly, in pts &lt; 50 yrs, 13% of stage IV patients were dMMR/MSI-H. Conclusions: Reflex testing of MMR/MSI status in CRC is feasible with concordant results. Routine testing results in identification of dMMR/MSI-H at or higher than published rates. Notable findings include the high prevalence in those with sporadic CRC and/or younger than 50 yrs. Continued impact analysis of this approach is warranted to maximize IO therapy offering.

Clinical and Genetic Implications of DNA Mismatch Repair Deficiency in Biliary Tract Cancers Associated with Lynch Syndrome.

Patients with Lynch syndrome (LS) have a significantly elevated lifetime risk of developing biliary tract cancers (BTCs) compared to the general population. However, few studies have characterized the clinical characteristics, genetic features, or long-term outcomes of mismatch-repair deficient (dMMR) cholangiocarcinomas associated with LS. A retrospective review of a prospectively maintained Familial High-Risk GI Cancer Clinic database identified all patients with BTCs evaluated from 2006 to 2016 who carried germline mutations in MLH1, MSH2, MSH6, or PMS2. Eleven patients with BTCs were identified: four perihilar, four intrahepatic, one extrahepatic, one gallbladder, and one ampulla of Vater. All patients had underlying germline mutations and a personal history of a LS-associated malignancy, most commonly (63.3%) colorectal cancer. Ten (90.9%) patients were surgically explored, and margin negative resection was possible in seven (63.3%). Chemotherapy (90.9%) and/or chemoradiation (45.5%) was administered to most patients. Among the seven patients presenting with non-metastatic disease who underwent surgical resection with curative intent, the 5-year overall survival rate was 53.3%. The median overall survival for the four patients not treated with curative intent was 17.2months. dMMR biliary tract cancers associated with LS are rare but long-term outcomes may be more favorable than contemporaneous cohorts of non-Lynch-associated cholangiocarcinomas. Given the emerging promise of immunotherapy for patients with dMMR malignancies, tumor testing for dMMR followed by confirmatory germline testing should be considered in patients with BTC and a personal history of other LS cancers.

Screening for mismatch repair deficiency in colorectal cancer: data from three academic medical centers.

Reflex immunohistochemistry (rIHC) for mismatch repair (MMR) protein expression can be used as a screening tool to detect Lynch Syndrome (LS). Increasingly the mismatch repair‐deficient (dMMR) phenotype has therapeutic implications. We investigated the pattern and consequence of testing for dMMR in three Irish Cancer Centres (CCs). CRC databases were analyzed from January 2005–December 2013. CC1 performs IHC upon physician request, CC2 implemented rIHC in November 2008, and CC3 has been performing rIHC since 2004. The number of eligible patients referred to clinical genetic services (CGS), and the number of LS patients per center was determined. 3906 patients were included over a 9‐year period. dMMR CRCs were found in 32/153 (21%) of patients at CC1 and 55/536 (10%) at CC2, accounting for 3% and 5% of the CRC population, respectively. At CC3, 182/1737 patients (10%) had dMMR CRCs (P < 0.001). Additional testing for the BRAF V600E mutation, was performed in 49 patients at CC3 prior to CGS referral, of which 29 were positive and considered sporadic CRC. Referrals to CGS were made in 66%, 33%, and 30% of eligible patients at CC1, CC2, and CC3, respectively. LS accounted for CRC in eight patients (0.8%) at CC1, eight patients (0.7%) at CC2, and 20 patients (1.2%) at CC3. Cascade testing of patients with dMMR CRC was not completed in 56%. Universal screening increases the detection of dMMR tumors and LS kindreds. Successful implementation of this approach requires adequate resources for appropriate downstream management of these patients.

198P Use of chemotherapy and mismatch repair deficiency testing in resected stage II colon cancer

Background Adjuvant chemotherapy for stage II colon cancer offers a small (2-3%) overall survival benefit and is not universally recommended. Currently, chemotherapy is often discussed with patients at increased recurrence risk based on histological criteria. Mismatch repair deficiency (dMMR) confers an improved prognosis and identifies patients who appear not to benefit from post-operative chemotherapy. Widespread immunohistochemistry testing in Australia began in 2013. Methods Prospective data were collected on all patients with resected stage II colon cancer between Feb-2010 to Feb-2015 across two large Australian hospitals. Rectal cancers were excluded. Data collected included patient demographics, tumour histology especially presence of high-risk features (<12 nodes, obstruction, perforation, poor differentiation, T4), dMMR testing frequency, chemotherapy use, and outcomes. Results 355 patients (56.1% female) were analysed. 25 (7.3%) received adjuvant chemotherapy (Cabrini 22/271 vs. Alfred 3/84,8% vs. 4%p = 0.22). Mean age was 77.9 years. No patient >80 years (52%) received chemotherapy. Presence of > 1 high-risk feature increased the likelihood of adjuvant chemotherapy. dMMR testing was performedon167patients(47%),mostoccurredpost-2013 (117/160vs. 50/195 patients).dMMRrateswere34.1%.2/57dMMRpatientsreceivedchemotherapy;both young(34and54years) withhigh-riskfeatures.28/355(7.8%) patientsrelapsed, including 3/25 (12%) who had received chemotherapy, and 25/330 (6.9%) who did not. 10/28(36%) ofrelapsesaredisease-freepost-resection,9(32%) alivewithmetastatic disease, and 9 deceased. An audit of familial cancer clinic referrals and BRAF mutation rate is underway. Conclusions >50%ofpatients were>80 years. Uptakeofchemotherapywaslow (7.3%).Despitethis,fewrelapseswereobserved(7.8%)and36%oftheseunderwent successful surgical salvage. dMMR testing should be performed in all stage II colon cancer patients being considered for adjuvant chemotherapy as it will identify a proportion(34.1%)ofpatientswhomaynotbenefitfromtreatment. Legal entity responsible for the study N/A Funding Alfred Hospital and Cabrini Health Disclosure All authors have declared no conflicts of interest.

Universal screening for Lynch syndrome (LS) in colorectal cancer (CRC) and survival.

415 Background: It has been argued that universal screening for LS should be performed on all patients (pts) diagnosed with CRC because screening LS pts and their relatives can detect early cancers and improve outcomes. This approach is being adopted by many comprehensive cancer centers around the country. The Ohio State University (OSU) began screening for deficient mismatch repair (dMMR) proteins by immunohistochemistry (IHC) on all CRC cases in 2006. This retrospective study aims to assess whether universal screening with IHC followed by conclusive testing (BRAF mutation in setting of MLH1 deficiency, and/or gene sequencing) affects pts survival. Methods: All CRC pts diagnosed at OSU from 3/2006-6/2012 with dMMR testing per IHC were included, and their charts were reviewed. Chi-square test was used to for categorical parameters, t-test was used for continuous variables, and the log-rank test for overall survival (OS) to compare the difference between pts with full conclusive testing vs those without full testing. Patients who expired within 4 weeks of diagnosis were excluded (n=10) from the OS analysis. Results: One hundred and twenty-two tumors (15.7% of 777 tumors tested) had dMMR by IHC during the study period. Fifty-one pts went on to have conclusive testing done; thirteen patients (25.5%) were diagnosed with LS, 29 patients (56.9%) had MLH1-hypermethylated tumors, and 9 patients (17.6%) were not found to have MMR mutations on gene sequencing. In an OS analysis, patients undergoing conclusive testing had significantly longer survival as compared to those with abnormal IHC without further testing (median OS not reached vs. 38 months, p=0.04). The difference was similar and remained marginally significant (p= 0.04-0.09) after correcting for age, stage and testing date using different models. Conclusions: Patients who had conclusive testing after initial abnormal universal screening with IHC had significantly longer OS compared to pts who had abnormal IHC without follow-up testing. It is possible that ascertainment bias and pts compliance influenced survival so larger prospective studies that document stage and treatment are needed to determine if increased surveillance in the setting of a LS diagnosis leads to better OS.

Use of Mismatch Repair Immunohistochemistry and Microsatellite Instability Testing

The mismatch repair (MMR) status of tumors is being increasingly recognized as a prognostic, predictive, and possible germline predisposition/Lynch syndrome (LS) biomarker in colorectal cancer and other cancer types, particularly in endometrial cancer. Current methods (clinical history and tumor morphology) to predict MMR deficiency (dMMR) are suboptimal, and implementation of reflex laboratory testing of appropriate tumors has been recommended, a strategy requiring test standardization and clinical coordination. Two web-based questionnaires were administered, a general and a specialist laboratory questionnaire, to establish the availability of such tests, requisite clinical/pathology integration, current mode of test initiation, reporting and recommendation practices, and education and attitudes among pathologists. Technical aspects were reviewed on the basis of specialist laboratory practice. Of 76 respondents, 21.5% were unaware or were uncertain whether they had access to MMR immunohistochemistry. Although 78.9% of respondents had access to such testing, an integrated approach to the identification of patients with LS is lacking, being limited to just 9 centers. The majority (70%) of testing is clinician initiated, with variable implementation of reflex testing and divergent practices in recommendation to test. Standardized reporting is lacking in many centers. Education on MMR in endometrial cancer is poor compared with that in colorectal cancer (P<0.0001). This multicenter questionnaire highlights heterogenous practices in dMMR testing and LS identification, both in clinical terms and with regard to technical aspects of testing. An integrated multidisciplinary approach is lacking, and there is a need to educate physicians and resolve ethical issues. A Canadian consensus statement and national guidelines on dMMR testing are urgently needed, requiring input from pathologists, clinicians, and genetic counselors.