Patients were stratified by hospital and surgeon. The study analyzed multiple factors, including age and gender, date of surgery, pathologic grade, pathologic stage, presence of MMR immunohistochemical (IHC) test, and presence of genetic counseling and testing for MMR-positive patients. Data was extracted from patient charts, pathology reports, and genetic reports. Only patients with primary adenocarcinomas were included in the study. In total, the study comprised 423 cases among the 5 hospitals. Results were tabulated and simple descriptive statistics were utilized to analyze the data.
Results
Of the 423 CRC patients treated at HHC during the study period, 45% were male and 55% were female, with an average age of 68.2 years (Table 1). The HHC Cancer Institute performed MMR IHC testing on 81.3% of all patients diagnosed in 2014 and 2015 (range, 30.8% to 94.5%). While the percentage of patients tested overall did not change from 2014 to 2015, it appreciably increased for the lower performing hospitals (Table 1). This improvement resulted from enhanced communication and establishment of pathology protocols for handling the tissue of patients with a cancer diagnosis.
Twenty-six (7.6%) of the 344 specimens tested were IHC abnormal, revealing a loss of 1 or more MMR gene products (Table 2). Of the patients with MMR-positive results, 15 (57.7%) received a genetic consult and 10 of these had a germline test of their MMR genes. Of note, 1 patient had been diagnosed with LS at an outside facility and therefore did not receive a genetic consult; 1 patient was unable to be reached for scheduling of a consult; 2 patients declined genetic testing; and 1 patient did not have their genetic test ordered.
Of the patients who underwent germline testing, 7 (70%) tested positive for LS (Table 2). Five LS patients tested positive for an MLH1 gene mutation, 1 tested positive for an MSH2 mutation, and 1 had a pathogenic variant of unknown significance (VUS) in their MLH1 gene.
The stage of cancer at diagnosis for MMR-negative, MMR-positive, and LS-positive groups was similar; nearly all patients were stage I, II, or III (Table 3). Compared to patients who were MMR-negative or MMR-positive, LS patients were younger (68.3, 60.9, and 47.6 years, respectively), and the majority were male (44.8%, 42.3%, and 57.1%, respectively).
Discussion
The shifting paradigm of health care delivery in America has led to increasing consolidation of hospitals into larger health care organizations. Consolidation creates a challenge when trying to implement a unified standard of care within distinct hospitals that comprise a health care system. In 2014, HHC integrated 2 additional hospitals into its system, for a total of 5 hospitals. As part of our quality improvement process, we wanted to explore the effect this had on universal MMR tumor screening for CRC patients among the 5 separate pathology departments, recognizing that implementation might take some time as protocols change. Although our Cancer Institute and Pathology Council had approved the universal MMR testing standard for all CCR patients, it was not clear that the standard had been embedded into pathology department standard practice.
The project reported here revealed substantial variance in MMR IHC testing among the 5 hospitals, suggesting the difficulty of implementing a unified standard of care among hospitals with separate groups of pathologists. This variance could result from several issues: lack of embedding the new standard in a series of steps to assure universal compliance; lack of agreement by pathologists on submitting every case; lack of follow-up by pathology staff to forward slides/tissue to the central lab for processing; and concern about privacy issues associated with conducting an unconsented genetic test.