Characteristics of HER2 FISH-Equivocal Breast Cancers and the Impacts on HER2 Status of 2018 ASCO/CAP Guideline

Background: According to 2018 ASCO/CAP guideline, HER2 FISH-equivocal breast cancers will be categorized as HER2 negative except those with HER2 IHC 3+. However, whether or not HER2 FISH-equivocal breast cancers was a heterogeneous group has not been well illustrated. Methods: 195 HER2 FISH-equivocal breast cancers were collected between 2014 and 2018. The molecular subtype was determined according to 2013 St Gallen consensus, and HER2 status was also redened following 2018 ASCO/CAP guideline. All cases were classied into 4 groups according to the average HER2 copy number (4.0-4.4, 4.5-4.9, 5.0-5.4, 5.5-5.9 signals/cell). The relationship between HER2 copy number and clinicopathological parameters was analyzed. Results: 183 (93.8%) of 195 FISH-equivocal cases were of luminal subtype, while the other 12 (6.2%) were undetermined. Using 2018 ASCO/CAP guideline, all FISH-equivocal cases were recategorized to HER2 negative. Therefore, 31(15.9%) cases were luminal A-like, 152 (77.9%) were luminal B-like (HER2 negative) and 12 (6.2%) were triple negative. The average HER2 copy number showed positive correlation with chromosome 17 polysomy, but had no signicant association with other clinicopathological parameters as well as prognosis. 17 (8.7%) cases were treated with trastuzumab, but showed no difference in prognosis with patients who didn’t receive targeted therapy. Conclusions: In this study, all HER2 FISH-equivocal breast cancers were reclassied to be HER2 negative according to 2018 ASCO/CAP guideline. Most of these patients were luminal B-like (HER2 negative). The average HER2 copy number had no signicant association with clinicopathological parameters, as well as prognosis.


Background
Breast cancer remains the leading cause of cancer incidence and mortality in women worldwide. Global cancer statistics estimated that there was about 2.1 million new cases in 2018, accounting for almost 25% of all female cancers cases [1].
Human epidermal growth factor receptor 2 (HER2) is found to be involved in several key cellular signal pathways, including proliferation, migration, and adhesion in breast cancers [2]. HER2-targeted drugs, such as Trastuzumab, which inhibit downstream signaling of these pathways, are very effective for HER2 positive breast cancer patients. Therefore, HER2 status is crucial in choosing therapeutic scheme.
Currently, HER2 testing mainly in two ways, immunohistochemistry (IHC) assays for protein overexpression and uorescence in situ hybridization (FISH) for gene ampli cation. Although the commonly adopted routine is to perform IHC assays initially, HER2 FISH is required when the IHC result is equivocal. HER2 status was determined according to the American Society of Clinical Oncology/College of American Pathologists (ASCO/CAP) guideline recommendations [3][4][5].
However, 2013 ASCO/CAP guideline caused an increased HER2 equivocal cases [6][7][8], whose HER2 status is unde ned. It is hard for medical oncologists to decide whether or not using HER2-targeted therapy for equivocal cases. In 2018, the guideline was updated and recommended a de nitive diagnosis for the former HER2 FISH-equivocal breast cancers, and almost all equivocal cases were categorized as HER2 negative except those with HER2 IHC 3+ [5]. Since the dual-probe FISH test gives an exact score of average HER2 copy number and the value ranges from 4.0 to 6.0. However, whether or not HER2 FISHequivocal breast cancers could be a heterogeneous group according to different HER2 copy number has not been well investigated.
In this study, we collected 195 HER2 FISH-equivocal invasive breast cancers diagnosed following 2013 ASCO/CAP guideline [4], and rede ned HER2 status according to 2018 updated guideline. The molecular subtype of those cases were categorized basing on different HER2 status. The relationship between HER2 copy number and clinicopathological parameters was analyzed. The impact of different HER2 copy number on prognosis was also illustrated.

Patients and study design
According to the 2013 ASCO/CAP guideline, 195 invasive breast cancers diagnosed to be HER2 FISHequivocal from 2014 to 2018 were retrieved from Shanghai Cancer Center. For all cases, IHC and FISH assays were performed on the same tissue block of surgical specimen. The diagnoses were reviewed by two experienced senior pathologists. Clinical characteristics of 195 study subjects are presented in Table 1. antibodies included ER, PR, HER2, Ki-67, E-cadherin, p120, AE1/AE3, calponin and p63. All of these antibodies were from Roche Ventana. Appropriate positive and negative controls were carried out simultaneously for all stains.
The status of ER and PR were determined following the criteria of ASCO/CAP guideline [10]. The ratio of nuclei-positive tumor cells to all tumor cells calculated visually represented the Ki-67 value, and a high Ki-67 expression was de ned as ≥ 20% [9]. E-cadherin and p120 were employed to discriminate lobular carcinomas from ductal carcinomas. Double immunostainings were performed with AE1/AE3/P63 or AE1/AE3/calponin to con rm the presence of invasive component.
A dual-probe FISH test using the PathVysion HER2 DNA probe Kit (Vysis Inc, Downers Grove, IL) was performed on the same specimen as IHC test following the manufacturer's instructions. FISH results were interpreted by two experienced pathologists independently.

Follow-up Information
Patients follow-up were carried out in Shanghai Cancer Center. The overall survival (OS) was computed from the date of surgery till death of any cause. Disease-free survival (DFS) was calculated from the date of surgery till the rst proven event including recurrence, distant metastasis, second primary tumor and death of any cause. The last follow-up was conducted in August 2018.

Statistical analysis
The correlation between clinicopathological factors and the average HER2 copy number was analyzed by chi-squared test or Fisher's exact test. OS and DFS were estimated with Kaplan-Meier curves analysis.
IBM SPSS Statistics software (version 21) was used to perform the statistical analysis. All P values were two-tailed, and difference with P value < 0.05 was considered to be statistically signi cant.
In term of histological grade, the collected samples were mainly distributed between Grade II and Grade III. The numbers of corresponding cases were 94 (48.2%) and 99 (50.8%), respectively. Only 2 (1%) cases were diagnosed as Grade I.   The correlation between those parameters and average HER2 copy number were analyzed. The results suggested that the average HER2 copy number showed positive correlation with the presence of chromosome 17 polysomy (P < 0.001). However, there were no signi cant correlation with other clinicopathological parameters.

Prognosis
The individualized therapy was decided through a multidisciplinary team meeting. All involved patients have taken surgery and treated with different combinations of chemotherapy, radiotherapy and hormonal therapy. In addition, among the 195 HER2 FISH-equivocal individuals, 17 (8.7%) have taken targeted therapy of trastuzumab preoperatively or postoperatively.
Patients follow-up data was collected in a long term and the most recent data was collected in August 2018. Among 195 study subjects, 16 patients who were out of contact, were excluded from survival analysis. The median duration of follow-up was 34 months (range 10-55). By the end of data collection, no recurrence of breast cancer had been seen, while distant metastasis had occurred in ve cases, including one lung metastasis, one liver metastasis, and three bone metastases. Besides, four individuals died without detail information.
Treatment response and prognosis of 17 patients taking HER2 targeted therapy were the major problem we concerned. Although two of them were treated with trastuzumab before surgery, they also received chemotherapy and/or radiotherapy. The Miller-Payne grading system was used to assess pathologic response to neoadjuvant chemotherapy [11], and the two patients were categorized as grade 3 and grade 4. The other 15 patients took targeted therapy after surgery, and no recurrence had been seen by the end of the follow-up.
Univariate analysis of the impact of prognostic factors on DFS and OS of breast cancer patients was performed. The results showed that the N (regional lymph nodes) stage signi cantly affected OS, and the M (distant metastasis) stage signi cantly affected both DFS and OS. However, the average HER2 copy number in FISH test had no signi cant impact on DFS and OS of breast cancer patients. Meanwhile, DFS and OS showed no signi cant difference between patients with and without targeted therapy (Table 3).  [12,13]. It is just what makes the determination of HER2 status a routine pathological examination for breast cancer patients worldwide [14].
IHC assay is performed for every breast cancer patient as routine, while FISH is an optional test. However, HER2 FISH is required when the IHC results is 2 + according to the ASCO/CAP guideline. Unfortunately, the result still can be equivocal even in HER2 FISH test, if the ratio of HER2 to CEP17 is < 2.0 with an average HER2 copy number between 4.0 and 6.0 [4]. It is hard for medical oncologists to make treatment decision for FISH equivocal patients. Therefore, an Expert Panel updated ASCO/CAP guideline to clarify HER2 equivocal status in 2018.
According to the new guideline, HER2 FISH-equivocal breast cancers will be categorized as HER2 negative except those with HER2 IHC 3+. As such, HER2 targeted therapy was not recommended to those patients.