563P - Cascade testing following universal immunohistochemistry for mismatch repair protiens

Date 08 October 2016
Event ESMO 2016 Congress
Session Poster Display
Topics Colon and Rectal Cancer
Pathology/Molecular Biology
Basic Scientific Principles
Presenter Nabeeha Karadawi
Citation Annals of Oncology (2016) 27 (6): 149-206. 10.1093/annonc/mdw370
Authors N. Karadawi1, G. O'Kane2, D. Gallagher3, S. Finn3, C. Muldoon4, N. Mulligan5
  • 1Medical Oncology, St James's Hospital, 8 - Dublin/IE
  • 2Medical Oncology, Princess Margaret Hospital, Toronto/CA
  • 3Cancer Molecular Diagnostics, University of Dublin Trinity College, D8 - Dublin/IE
  • 4Histopathology Department, St James's Hospital, Dublin/IE
  • 5Histopathology Department, Mater Misericordiae University Hospital University College Dublin, Dublin/IE



Reflex immunohistochemistry (rIHC) to detect mismatch repair deficiency (dMMR) is a recommended screening tool for the detection of Lynch Syndrome (LS) in incident colorectal cancers (CRCs). Additional BRAF testing in tumours exhibiting MLH1 loss can help exclude sporadic cases, which do not require CGS. IHC is increasingly performed as the dMMR phenotype may be considered both a prognostic and predictive biomarker. The downstream management of dMMR results can be complex and patients are not always referred to clinical genetic services (CGS).


We investigated the work-up of dMMR CRC patients detected between 2005-2013 at two academic institutes. During this period centre 1 performed IHC only at physician request and centre 2 implemented rIHC in November 2008. Neither performed additional BRAF testing prior to CGS referral. Patients who did not receive a CGS referral were identified. Ethical approval was obtained and retrospective BRAF testing was performed on tissue from patients who were still alive and whose tumours exhibited loss of MLH1.


During the studied period, 2169 new CRC patients were reviewed. 87 (4%) of these were identified as dMMR CRCs. 39(45%) patients were referred and investigated for LS. Forty-seven patients were not referred. The characteristics of these patients are shown in table 1. 41 of 47 (87%) patients had tumours with evidence of MLH1 loss, 35 of these were older than 70 years. Nineteen patients had died and BRAF testing is on-going in this cohort. 14 had a BRAFV600E mutation and 8 were wild-type. Only 2 of these were less than 70 years. These patients will now be offered genetic counselling.

N (47) %
Median Age (yrs.) 78yrs (55-90)
MMR protein loss
MLH1/PMS2 41
>70yrs 35
≤70yrs 6
MSH6 2
PMS2 1


Over half of identified patients were not referred to clinical genetic services. Braf testing can enrich the appropriate patients requiring referral and thus improve the efficiency of screening. Certain patients with BRAF-wild type tumours require additional work-up.

Clinical trial identification

Legal entity responsible for the study

St James Hospital




All authors have declared no conflicts of interest.