Clinical significance of HER2 and EGFR expression in colorectal cancer patients with ovarian metastasis

A total of 31cases of CRC ovarian metastases from the Department of Gastrointestinal Surgery, Guangxi Medical University Affiliated Cancer Hospital from June 2013 to December 2017 were selected. The average age was 50.0 ± 11.0 years. There were 14 cases of synchronous metastases, 17 cases of metachronous metastases, 12 cases of unilateral ovarian metastases, and 19 cases of bilateral metastases; 20 cases had paired primary lesions and ovarian metastases, 9 cases had ovarian metastases only, and 2 cases had primary tumours only. The inclusion criteria were as follows: (1) postoperative diagnosis of primary CRC based on histopathology; and (2) ovarian metastases confirmed by postoperative pathology. Besides histological analysis, expressions of CK7, CA125, CK20, CEA, CDX2, PAX8 were also determined by immunohistochemistry to confirm ovarian metastasis when necessary. The exclusion criterion was the presence of primary ovarian cancer or ovarian metastases derived from other cancers. At the same time, retrospective analysis of third-stage patients with non-ovarian metastases who were treated at our hospital as the control group were selected as controls for EGFR or HER2. The inclusion criteria for the controls were (1) female patients with primary colorectal tumours confirmed by postoperative histopathology; (2) no ovarian metastases detected by computed tomography; and (3) previous confirmation of the HER2 or EGFR expression status with immunohistochemistry (IHC). In total, 22 patients were included in the HER2 control group, with an average age of 65.4 ± 14.5 years; four cases had liver metastases, one case had bladder metastasis, two cases had peritoneal metastasis, and 15 cases had tumour deposits in perirectal/mesorectal adipose tissues. The EGFR control group included 24 patients, with an average age of 67.6 ± 11.9 years, six cases had liver metastases, one had lung metastases, two had bone metastases (1 combined with lung metastases), one had bladder metastases, two had abdominal metastases, and 13 cases had tumour deposits in perirectal/mesorectal adipose tissues. The study was approved by the Ethics Committee of the Affiliated Cancer Hospital of Guangxi Medical University, and all patients signed informed consent.

Surgical specimens were fixed in 10% neutral formaldehyde solution, dehydrated, embedded in paraffin, and serially sectioned at a thickness of 4 μm. Serial sections from the same blocks were used for HER2 and EGFR analysis. Specimen sections were dewaxed by xylene, rehydrated with a standard ethanol gradient, and subjected to high temperature and high pressure for antigen retrieval. The slices were incubated in H₂O₂ for 10 min at room temperature, subjected to dropwise addition of the corresponding primary antibody followed by incubation at 4 °C overnight, rinsed with phosphate buffered saline (PBS), and subjected to dropwise addition of secondary antibody. A DAB solution was added to visualize the antibody binding, after which the sections were rinsed with distilled water, counterstained with haematoxylin, dehydrated with an ethanol gradient, and fixed with xylene and gelatin. Mouse anti-human HER2/c-erB-2 monoclonal antibody (product number: ZM-0065, clone number: UMAB36) and mouse anti-human EGFR monoclonal antibody (product number: ZM-0093, clone number: UMAB95) were purchased from Beijing Zhongshan Jinqiao Company. A ready-to-use rapid IHC MaxVision™² detection kit (cat. KIT-5920) was purchased from Fuzhou Maixin Co., Ltd. The experimental procedure was carried out according to the manufacturer’s instructions. The negative control slices were incubated with PBS instead of primary antibody.

HER2 is mostly localized to the cell membrane, while EGFR is expressed in the cell membrane and cytoplasm. HER2 expression was scored as described in the VENTANA study [13]: (−), no tumour cell staining or less than 10% cell membrane staining; (+), > 10% cells had weak or almost undetectable cell membrane staining; (++), > 10% cells were weak to moderate cell membrane staining; and (+++), > 10% of cells have strong cell membrane staining. EGFR expression was determined according to the staining intensity and the number of positive cells. The staining intensity was scored as follows: no staining, 0 points; staining faintly visible, 1 point; weak-medium staining, 2 points; and strong staining, 3 points. The number of positively stained cells was scored as follows: ≤ 10% positive cells, 1 point; 11 to 50%, 2 points; 51 to 80%, 3 points, and ≥ 80%, 4 points. The total score was calculated by multiplying the staining score by the positive cell number score. The expression of EGFR was determined as follows: 0 to 2 points, (−); 3 to 5 points, (+); 6 to 8 points, (++); and 9 to 12 points, (+++). For both HER2 and EGFR, (−) and (+) were considered negative, whereas (++) and (+++) were considered positive. All the results were independently assessed by two highly experienced pathologists in a double-blinded manner and agreed upon by consensus.

All patients in this group were followed up mainly by outpatient review and telephone. The recorded survival time was from the time of diagnosis of ovarian metastases to the time of death or the last follow-up (August 2018).

HER2 is mainly expressed in the cell membrane (Fig. 1 a-d). The primary tumour HER2-positive rate in ovarian metastatic CRC patients was 54.5%, which was significantly higher than that in HER2 control patients (36.4%, P < 0.05) (Table 1). There was no significant correlation in the expression of HER2 with age, tumour site, tumour differentiation, tumour diameter, number of positive mesenteric lymph nodes or vascular cancer embolus (P > 0.05). There was no significant correlation between the expression of HER2 and the clinicopathological features in CRC patients with non-ovarian metastasis (P > 0.05) (see Table 2).

EGFR was localized to the cell membrane and cytoplasm (Fig. 1e-h). There was no significant difference in the expression of EGFR between CRC patients with ovarian metastasis and those with non-ovarian metastasis (P > 0.05, Table 2). The positive rate of EGFR expression in the CRC ovarian metastasis group was 63.6%. No significant difference was found between the expression of EGFR and the clinicopathological variables (P > 0.05, Table 3).

The positive rate of HER2 in ovarian metastases was 44.8% (13/29). The expression of HER2 in ovarian metastases was associated with peritoneal metastasis and bilateral ovarian infiltration (P < 0.05, Table 4). HER2 expression in ovarian metastases showed no significant relationship with age, synchronous or metachronous ovarian metastasis, or primary site (P > 0.05, Table 4).

The positive expression rate of EGFR in ovarian metastases was 65.5% (19/29). There was no significant difference in EGFR expression between ovarian metastases and the clinicopathological features (P > 0.05, Table 4).

In paired primary tumours and ovarian metastases samples, 5 out of 20 cases were HER2 positive in both primary tumours and ovarian metastases, 4 cases were HER2 positive in only primary tumours, and 5 cases were only positive in ovarian metastases, 9 cases were EGFR positive in both primary tumours and ovarian metastases, 5 cases were EGFR positive in only primary tumours, and 3 cases were only positive in ovarian metastases. The EGFR-positive rate was 69.0% in ovarian metastases, which was concordant with the 63.6% observed in the matched primary tumours (Tables 1 and 4).

Among all patients with ovarian metastases, the median survival time in metastases HER2-negative patients was 32.0 ± 8.3 months [CI 95%: 15.80–48.19], compared to 17.0 ± 5.2 months [CI 95%:6.76–27.24] in HER2-positive patients (Fig. 2).