Breast Biomarker Cytology Practice in 2023: Results of a College of American Pathologists Survey

Breast cancer is the most common malignancy in women and the second leading cause of cancer-related death among women.¹ Estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) are critical biomarkers for both therapeutic decision-making and prognosis in primary as well as recurrent/metastatic breast cancer management. The American Society of Clinical Oncology/College of American Pathologists (ASCO/CAP) guidelines^2,3 recommend testing ER, PR, and HER2 in tissue specimens obtained from recurrent or metastatic lesions in patients with breast cancer because of the potential for receptor status to change during the progression from primary to recurrent tumors.^4–8

However, obtaining surgical specimens from some metastatic sites can be challenging, and cytologic specimens are often the only available option. Cytology has proven to be a minimally invasive and cost-effective method for diagnosing metastatic breast carcinoma, especially in anatomic sites such as body cavities and parabronchial lymph nodes, where fine-needle aspiration (FNA) is often the most feasible approach for obtaining diagnostic material. Breast biomarker testing has been validated for use on cytologic specimens.^9,10 Some studies have reported that receptor testing on breast cancer cytology specimens demonstrates good agreement with corresponding tissue specimens for hormone receptors.^11–15 Applying the ASCO/CAP guideline of 1% cutoff for ER and PR positivity, studies have demonstrated concordance rates of 80% to 99%.^10,14,16,17 For HER2 testing, agreement with tissue specimens is reportedly improved when dual in situ hybridization is performed on cases with HER2 2⁺ immunostaining results.^18–20 Additionally, HER2 immunocytochemistry performed on cell blocks has shown correlation rates of 98% or greater with HER2 fluorescence in situ hybridization, as reported by Vohra et al.¹⁰

Despite these advancements, data regarding breast biomarker cytology practices across different laboratories remain limited. The College of American Pathologists (CAP) has used survey questionnaires to assess practice changes, identify emerging trends, and establish national metrics in various cytologic areas. The 2023 CAP Breast Biomarker Cytology Practice Supplemental Questionnaire aimed to evaluate current breast biomarker cytology practices and trends, identify differences among institution types and between domestic and international laboratories, and provide baseline data for future studies in diverse practice settings.

METHODS

RESULTS

Survey Results

Of the 1914 laboratories that participated in the D-mailing of the CAP Nongynecologic Cytopathology 2023 survey, 728 laboratories (38.0%) responded to question 1 of the Breast Biomarkers Supplemental Questionnaire.

Among the 728 respondents, 180 laboratories (24.7%) performed breast biomarker testing on FNA/other cytologic specimens for metastatic breast cancer, including 161 US laboratories (89.4%) and 19 international laboratories (10.6%). Among the 167 laboratories that specified their institution types, 107 (64.1%) were nonacademic hospitals/medical centers, 24 (14.4%) were academic hospitals/medical centers, 24 (14.4%) were nonhospital clinics/laboratories, 7 (4.2%) were national/corporate laboratories, and 5 (3.0%) were veterans hospitals/DOD (Table 1).

Table 1.Summary of Survey Results of Breast Biomarker Cytology Practices

View Fullscreen

Specimen Volumes

Among the 167 respondents who specified nongynecologic cytology specimen volume, most laboratories (43.7%; 73 of 167) reported a volume between 1001 and 4000 cases, followed by volumes between 501 and 1000 (23.4%; 39 of 167), more than 4000 (19.8%; 33 of 167), and 500 or fewer (13.2%; 22 of 167).

Breast Biomarker Tests

Among the 177 respondents who specified the breast biomarkers tested in their laboratory, ER/PR was the most evaluated test biomarker (98.9%; 175 of 177), followed by HER2 (93.2%; 165 of 177), Ki-67 (48.6%; 86 of 177), programmed death ligand-1 (PD-L1) (20.9%; 37 of 177), and mismatch repair (MMR) (19.8%; 35 of 177).

Specimen Types

Among the 156 respondents who specified the specimen types validated for breast biomarker testing, FNA was the most validated specimen type (85.3%; 133 of 156), followed by body cavity fluids (82.1%; 128 of 156) and others (3.2%; 5 of 156).

Specimen Preparation Methods

All 165 respondents validated cell blocks for breast biomarker tests, with a few laboratories also validating cytospin slides (3.0%; 5 of 156), liquid-based slides (3.0%; 5 of 156), air-dried direct smears (2.4%; 4 of 156), touch preparations (2.4%; 4 of 156), ethanol-fixed direct smears (1.8%; 3 of 156), and destained slides (1.8%; 3 of 156).

Collection Media

For collection media, CytoLyt was the most used collection medium (55.4%; 87 of 157), followed by balanced salt solution (16.6%; 26 of 157), Roswell Park Memorial Institute Medium (RPMI) (13.4%; 21 of 157), and CytoRich Red (9.6%; 15 of 157). However, 31 laboratories reported using collection media other than those specified.

Fixative

Among the 167 laboratories that responded, almost all indicated using formalin (90.4%; 151 of 167), while a few laboratories used other fixatives: 95% ethanol (5.4%; 9 of 167) and methanol-based fixative (3.0%; 5 of 167).

Primary HER2 Tests (Initially Performed HER2 Test)

In-house HER2 immunohistochemistry (IHC) was the most frequently used primary HER2 test (71.6%; 116 of 162), followed by send-out HER2 IHC (23.5%; 38 of 162), in-house HER2 in situ hybridization (ISH) (3.7%; 6 of 162), and send-out HER2 ISH (1.2%; 2 of 162).

Sign-out Practice

General pathologists primarily interpreted breast biomarkers on cytologic specimens in 68.6% (118 of 172) of the laboratories that responded, while cytopathologists primarily interpreted biomarkers in 39.0% (67 of 172) of laboratories. Additionally, breast pathologists signed out breast biomarkers on cytology specimens in 27.3% (47 of 172) of laboratories and molecular pathologists in 7.0% (12 of 172) of laboratories.

For HER2 ISH tests on cytologic specimens, general pathologists primarily interpreted them in 56.3% (58 of 103) of the laboratories that responded, while cytopathologists interpreted them in 33.0% (34 of 103) of laboratories, breast pathologists in 31.1% (32 of 103) of laboratories, and molecular pathologists in 19.4% (20 of 103) of laboratories.

Other Information Related to Breast Biomarkers From Cytologic Specimens

In reporting breast biomarker results, 75.7% (128 of 169) of laboratories reported fixative information, 65.7% (111 of 169) reported fixation time, and 51.5% (87 of 169) reported cold ischemic time. Slightly more than half (55.8%; 96 of 172) of the responding laboratories indicated that proficiency testing was required for pathologists interpreting breast biomarkers on cytologic specimens. Digital imaging platforms were used by only 12.9% (22 of 171) of responding laboratories. Forty-four laboratories (28.0%; 44 of 157) required cellularity adequacy for interpreting breast biomarkers on cytologic specimens, while 113 laboratories did not. Among the laboratories that required cellularity adequacy, most specified a requirement of more than 100 cells (50%; 22 of 44), followed by 20 to 50 cells (22.7%; 10 of 44), 51 to 100 cells (18.2%; 8 of 44), and fewer than 20 cells (9.1%; 4 of 44).

Breast Biomarker Reporting

Seventy-five percent (127 of 169) of laboratories indicated that they used a template to report breast biomarker results, and 89.1% (139 of 156) of laboratories reported HER2 IHC scores of 0 or 1⁺ for HER2-negative cytology cases.

Comparison of Breast Biomarker Cytology Practice Among Institution Types

Among the 604 participants that provided all institution demographic information, 159 laboratories performed breast biomarker testing on FNA/other cytologic specimens for metastatic breast cancer. The survey results identified the following statistically significant differences (Table 2).

Table 2.Summary of Specific Breast Biomarker Cytology Practices With Statistically Significant Differences Between Institution Types

View Fullscreen

More than half (57.5%; 23 of 40) of academic hospitals reported performing breast biomarker testing on cytologic specimens when compared to nonacademic hospitals (26.4%, 101 of 383; P = .04), nonhospital clinics/laboratories (22.1%, 23 of 104; P = .003), and national/corporate laboratories (14.6%, 7 of 48; P = .001).

Institution type was also associated with the personnel interpreting HER2 ISH testing (P = .001); however, the sample size was insufficient for testing pairwise comparisons. General surgical pathologists were more likely to interpret HER2 ISH test results in nonacademic hospitals than in academic hospitals (67.8% [40 of 59] versus 15.4% [2 of 13]), while a higher percentage of academic hospitals reported HER2 ISH interpretation by molecular pathologists than did nonacademic hospitals (30.8% [4 of 13] versus 8.5% [5 of 59]).

Comparison of Breast Biomarker Cytology Practice Between Domestic and International Laboratories

Among the 604 responding laboratories, 491 were domestic (United States), while 113 were international. We compared breast biomarker cytology practices between domestic and international laboratories, and the following statistically significant differences were identified (Table 3).

Table 3.Summary of Specific Breast Biomarker Cytology Practices With Statistically Significant Differences Between Domestic and International Laboratories

View Fullscreen

Significantly more domestic laboratories (28.9%; 142 of 491) reported performing breast biomarker testing in cytologic specimens than international laboratories (15.0%; 17 of 113) (P = .001). For specimen types, domestic laboratories reported performing breast biomarker testing on FNA specimens more often than international laboratories (87.8% [122 of 139] versus 64.7% [11 of 17]; P = .02). Cold ischemic time was more frequently reported in breast biomarker results from domestic laboratories than international laboratories (55.0% [77 of 140] versus 29.4% [5 of 17]; P = .05).

DISCUSSION

Cytology has proven to be a minimally invasive and cost-effective method for diagnosing recurrent/metastatic breast carcinoma in anatomic sites where surgical specimens are difficult to obtain. In these instances, cytology material is the only specimen available for the recommended biomarker (hormone receptor and HER2) testing in recurrent/metastatic breast carcinoma. Breast biomarker testing has been validated in cytologic specimens,^9,10 and studies have demonstrated good agreement between breast biomarker results from cytologic specimens and surgical specimens.^10–20 However, the current survey revealed that only approximately 25% of 728 responding laboratories routinely evaluated breast biomarkers in cytologic specimens in 2023, suggesting that breast biomarker testing was not widely used in routine cytopathology practice, which potentially raises a concern for optimal patient care. Although the biomarker status is available in tissue samples at the initial diagnosis of invasive breast carcinoma, there exists a likelihood for altered biomarker status in the subsequent recurrent/metastatic disease.

It is not surprising that ER/PR (98.9%) and HER2 (93.2%) were the top breast biomarkers evaluated in cytology specimens among all responding laboratories from our survey, given their well-established predictive and prognostic utility for metastatic breast cancer. Furthermore, our survey results revealed that 48.6% of respondent laboratories tested Ki-67 and approximately 20.0% of laboratories tested PD-L1 or MMR in cytologic specimens in 2023, suggesting the recognition and rapid adoption of these tests for newly emerged, targeted therapies in specific clinical settings.^21–24

Current survey results revealed the most common cytologic specimen types validated for breast biomarker testing were FNA (85.3%) and body cavity fluid (82.1%). All respondents validated cell block for breast biomarker testing, indicating the importance and appropriateness of cell block for ancillary studies in cytology practice.

Our survey revealed most laboratories used CytoLyt as a collection medium (55.4%) to collect materials, followed by balanced salt solution (16.6%), RPMI (13.4%), and CytoRich Red (9.6%). In addition, the current survey found formalin was the predominant fixative (90.4%), consistent with breast biomarker guideline recommendations.^2,3 Very few laboratories used other fixatives including ethanol (5.4%) and methanol (3.0%). Maleki et al²⁵ assessed ethanol-fixed cell blocks for breast biomarkers and found that ethanol fixation did not affect ER or PR results but produced a higher rate of variability for HER2 results. However, a recent study found that alcohol fixation did not reveal a clinically significant impact on ER or HER2 IHC results for cell-line materials across multiple laboratories using different antibodies.²⁶ Accurate evaluation of breast biomarker results depends on preanalytic variables including cold ischemia time and fixation time.^27–32 When dealing with tissue samples, laboratories have educated the clinicians and established workflows in an effort to comply with the requirements for optimal cold ischemia and fixation times specified in the ASCO/CAP guidelines.^2,3 Fixation time refers to the amount of time that the tissue spends in fixative, which can be obtained and recorded within cytologic laboratories. However, cold ischemia time involves clinical staff and is not always available for cytologic specimens, especially when dealing with body fluids. Indeed, the current survey revealed that 65.7% (111 of 169) of laboratories reported fixation time, while only 51.5% (87 of 169) reported cold ischemic time. The CAP Laboratory Accreditation Program standard CYP.4550 pertains to the cytology laboratories’ compliance with monitoring of cold ischemia and fixation times for ER and HER2 assessments by IHC, and it is important for laboratories to be aware of the requirement and implement that practice. Laboratories that perform breast biomarker testing need to participate in proficiency testing according to the CAP Laboratory Accreditation Program’s requirement. However, only slightly more than half (55.8%; 96 of 172) of surveyed laboratories indicated that proficiency testing was required for pathologists interpreting breast biomarkers on cytologic specimens. CAP checklists require pathologists interpreting breast biomarkers to have at least an alternative assessment if not proficiency testing.

Digital imaging analysis has been used to quantify breast biomarkers on surgical specimens more frequently than on cytologic specimens.^33–43 Our survey revealed digital imaging platforms were used by only 12.9% (22 of 171) of responding laboratories, consistent with previous reports.

When laboratories were stratified by their institution types, some significant differences were identified. More academic laboratories reported performing breast biomarker testing on cytologic specimens than did nonacademic laboratories. Additionally, a higher percentage of academic hospitals reported HER2 ISH interpretation by molecular pathologists than did nonacademic hospitals, indicating more subspecialized practice in academic settings.

Some significant differences in breast biomarker testing in cytologic specimens existed between US domestic and international laboratories. First, significantly more domestic laboratories reported performing breast biomarker testing in cytologic specimens than did international laboratories. Second, domestic laboratories reported performing breast biomarker testing on FNA specimens more often than international laboratories. Lastly, cold ischemic time was more frequently reported in breast biomarker results from domestic laboratories than from international laboratories, indicating more adoption of CAP guidelines in domestic laboratories.^2,3

CONCLUSIONS

This is the first survey by the CAP to investigate breast biomarker cytology practices among participating domestic and international laboratories. Our study identified some differences in practices among institution types and between domestic and international laboratories. More importantly, the results provide a valuable baseline to support further studies and/or guidelines to promote and refine these practices.