Effect of mismatch repair protein status on outcomes in high intermediate risk endometrial cancer

Abstract

<h3>Objectives:</h3> To compare recurrence rates between patients with mismatch repair proficient (pMMR) and MMR deficient (dMMR) high intermediate risk (HIR) endometrial cancer <h3>Methods:</h3> Adult women with HIR endometrial cancer diagnosed between January 1, 2016, and December 31, 2018, were identified from UC San Diego and Kaiser Southern California. Patient and disease characteristics were collected by chart review. Cases were limited to those who underwent hysterectomy with endometrioid histology and stage 1 disease. HIR disease was defined as the presence of three risk factors if age < 50 years, two risk factors if age ≥50 and <70 years, and one risk factor if age ≥70 years. Risk factors included tumor grade 2 or 3, the presence of lymphovascular space invasion (LVSI), and ≥50% myometrial invasion as outlined in GOG protocol 249. Categorical variables were compared using Chi-square tests, and continuous variables were compared using Student's t-test. The Kaplan-Meier method was used to evaluate time to recurrence. Assuming a two-sided alpha error of 0.05, power calculation identified a sample size of 174 would be needed to detect a 0.15 difference in recurrence rates between cohorts with 80% power. <h3>Results:</h3> Two hundred forty-four patients with HIR endometrial cancer were identified during the study period. One hundred fifty-eight patients (65%) were pMMR by immunohistochemical staining, and 86 patients (35%) were dMMR. There were no differences in mean age, mean body mass index (BMI), race, grade, mean follow-up time, or type of adjuvant treatment received between the pMMR versus dMMR cohorts. In total, 40 (16%) patients experienced cancer recurrence, and there were 11 (4.5%) cancer-related deaths during the study period. There was no statistically significant difference in recurrence rates between the pMMR and dMMR cohorts (15% versus 19%, p=0.49). The lack of association between MMR status and recurrence persisted after controlling for depth of myometrial invasion and LVSI in logistic regression multivariable analysis. pMMR tumors were more likely to exhibit ≥50% myometrial invasion than patients with dMMR tumors (p=0.03). dMMR tumors were significantly more likely to exhibit LVSI than pMMR tumors (p<0.001). There was no statistically significant association between the type of adjuvant treatment received and MMR status among those who recurred (p=0.58). Among patients who recurred, site of first recurrence (local vs distant) was not associated with MMR status (p=.90). The mean time to recurrence was 531 days in the pMMR cohort (n=24) and 473 days in the dMMR cohort (n=16), which was not significantly different (p=0.28). <h3>Conclusions:</h3> Our analysis represents one of the largest recurrence rate analyses of HIR endometrial cancer as stratified by MMR status. Although our pMMR and dMMR patient cohorts with HIR did not differ with respect to patient characteristics, dMMR tumors were significantly more likely to have <50% myometrial invasion and to exhibit LVSI. Despite these differences, we observed no difference in recurrence rates or time to recurrence between the two groups, which is in contrast to previously published reports of better prognosis among patients with pMMR tumors. Further study is ongoing to evaluate the effect of mismatch repair status on treatment selection and patient outcomes.