This clinical case review aimed to identify phenotypic variations in colorectal and extracolonic cancer expression between hereditary nonpolyposis colorectal cancer (HNPCC) families with MLH1 and MSH2 germline mutations and the general population. Colorectal cancer onset and site distribution were compared among 67 members of MLH1 kindreds, 45 members of MSH2 kindreds, and 1,189 patients from the general population. Synchronous and metachronous cancer rates, tumor stage, extracolonic cancer incidence, and survival were also compared. Mean ages of colorectal cancer onset were 44, 46, and 69 years for MLH1, MSH2, and the general population, respectively (P < 0.001). More proximal and fewer distal colon cancers were noted in HNPCC than the general population (P < 0.001, P = 0.04). Site distribution showed disparity of rectal cancers (8 percent MLH1 vs. 28 percent MSH2; P = 0.01) based on genotypes. Overall, synchronous colorectal cancer rates were 7.4, 6.7, and 2.4 percent for MLH1, MSH2, and the general population, respectively (P = 0.016). Annual metachronous colorectal cancer rates were 2.1, 1.7, and 0.33 percent for MLH1, MSH2, and the general population, respectively (P = 0.041). Colorectal cancer stage presentation was lower in HNPCC than the general population (P = 0.0028). Extracolonic cancers were noted in 33 percent of MSH2 patients, compared with 12 percent of MLH1 patients and 7.3 percent of the general population with colorectal cancers (P < 0.001). Combined MLH1 and MSH2 ten-year survival was 68.7 percent compared with 47.8 percent for the general population (P = 0.009 stage stratified, hazard ratio 0.57). The presence of rectal cancer should not preclude the diagnosis of HNPCC, because the incidence of rectal cancer in MSH2 was comparable with that in the general population. Phenotypic variations, including the preponderance of extracolonic cancers in MSH2 patients, did not result in survival differences between genotypic subgroups. These phenotypic features of HNPCC genotypes may have clinical significance in the design of specific screening, surveillance, and follow-up for affected individuals.
Read full abstract