A cytohistological correlation: Proliferative breast disease with atypia

INTRODUCTION

Breast fine-needle aspiration cytology (FNAC) is a rapid, cost-effective, and minimally invasive diagnostic procedure.^[1] Despite the increasing prevalence of breast core-needle biopsy (CNB), some centers continue to utilize FNAC as the primary diagnostic tool. Although CNB can assess invasion and determine hormone receptor status of the tumor, it is a time-consuming and expensive test with potential complications. The evaluation method, known as the “Triple Test,” which incorporates FNAC results, radiological and clinical interpretations, has an almost absolute positive predictive value.^[2,3]

Cytological criteria for grading and interpretation of breast FNAC specimens were introduced by Masood et al.^[4] The grading system had four categories as follows: Non-proliferative breast disease, proliferative breast disease without atypia, proliferative breast disease with atypia (PBDA), and carcinoma in situ or invasive cancer. The grading system takes cellular arrangement, cellular and nuclear pleomorphism, the status of myoepithelial cells, nucleoli, and the texture of chromatin into consideration.^[4] Over the years, a modified version of this grading system was proposed and accepted prevalently.^[5-7] PBDA is an important diagnostic category for the cytopathologists.^[4-12] Cytopathological diagnosis of PBDA supported by relevant clinical and radiological data provides invaluable information, easing the treatment burden of the patient.^[12]

In May 2016, at the Yokohama International Congress of Cytology, a group of cytopathologists, radiologists, surgeons, and oncologists, all of whom were experts in the management of breast lesions, initiated the development of the International Academy of Cytology (IAC) system for reporting breast fine-needle aspiration biopsy (FNAB) Cytology.^[13] Following the publication of a succinct overview of the proposed system, the atlas of the IAC Yokohama System for Breast FNAB Cytology was made accessible for global utilization.^[1,14] The Yokohama system is founded on the principles of cytomorphology and establishes diagnostic criteria for a spectrum of breast lesions. The system employs five clearly defined categories: Insufficient/inadequate, benign, atypical, suspicious of malignancy, and malignant.^[14] Since the global use of the Yokohama system in the field of breast cytology, researches and meta-analyses have been conducted on its reproducibility.^[15-26] The atypical category in the Yokohama system predicts a risk of malignancy (ROM) varying between 13% and 15.7%.^[14] In contrast, the studies using Yokohama system show higher ROM (25–37.5%) in the atypical category.^[6,15,17-26] Atypical category was scrutinized due to its heterogeneity.^[16] On the other hand, the suspicious category predicts a ROM ranging between 84.6% and 97.1%.^[14]

Based on FNAC criteria, PBDA is described as areas with somewhat disordered cellular arrangement and mild cytologic features of a benign-looking aspirate.^{[4,5,12,27,28]} Proliferative breast lesions with atypia are generally assigned either to the atypia or to the suspicious categories.^[12,14] Given that the Yokohama system is not routinely used in our center, the term of proliferative breast lesion with atypia is either used as the main result or mentioned in the interpretation of the results. There are limited studies with a scarce number of cases on PBDA.^{[4,5,9,10,12,28,29]} In this study, we wanted to address the clinicopathological importance of the result reported as “proliferative breast disease with atypia.”

MATERIAL AND METHODS

A computer-based search of all reports of breast FNAC cases signed out by two cytopathologists at the Izmir Katip Celebi University Ataturk Training and Research Hospital from January 2011 through September 2020 was carried out. The pathologists involved in this research received a cytopathology fellowship training program and were routinely examining breast FNAC specimens. Breast FNAC specimens reported as proliferative breast lesions with atypia were retrieved from the hospital data files. The cytological diagnosis of a proliferative breast lesion with atypia was established in accordance with the grading criteria proposed by Masood et al. and Nandini et al.^[4,5] This grading system is based on the assessment of the following parameters: cellular arrangement, cellular and nuclear pleomorphism, the status of myoepithelial cells and nucleoli, and the texture of chromatin. The system rates these parameters with scores ranging from 1 to 4, with four representing the most severe and one the least severe manifestations. The diagnosis of PBDA is established when the assigned scores range between 15 and 18.^[4,5] From 2011 to 2016, two Pap-stained slides and two May-Grünwald Giemsa-stained slides of each case were evaluated. The laboratory started using liquid-based cytology in 2017. Since then, in most instances, two Pap-stained slides, two May-Grünwald Giemsa-stained slides, and a Pap-stained slide from liquid-based cytology were evaluated. Liquid-based cytology slides were prepared using either the SurePath^® or ThinPrep^® techniques. Alcohol fixed slides were Pap-stained by an automated stainer. Air-dried slides were manually stained with May-Grünwald Giemsa stain produced by the Istanbul (Turkey) branch of the pharmaceutical company Merck Sharp and Dohme^®.

RESULTS

A total of 3125 breast FNAC specimens were identified, and 3.4% (n = 107) of them (including specimens from only two male patients) received the diagnosis of PBDA. The mean age of the patients with PBDA was 41.3 years. Among them, 39 (36.4%) cases were not followed up for further investigation. While 21 (19.6%) patients were lost to follow-up, 16 (15.0%) cases were deemed benign based on the triple test results, and 2 (1.8%) patients declined to undergo a second diagnostic procedure. Both male patients were lost to follow-up. Sixty-eight cases with histological correlation underwent excisional biopsy (n = 35: 32.7%), partial mastectomy (n = 28: 26.2%), and only breast CNB (n = 5: 4.7%). Five cases that underwent a CNB received the diagnosis of invasive breast carcinoma (n = 3) or fibroadenoma (FA) (n = 2). The breast cancer cases were not followed up with any further investigation in the institution. The general features of the cases are summarized in Table 1.

A total of 68 cases were included in this cytohistological correlative study. The ROM was 44.1%. The histological correlation of proliferative breast lesions with atypia results is demonstrated in Table 2. Histopathologically, all cases except one with invasive or microinvasive breast carcinomas (38.2%) were Nottingham grade 1 or 2 cancers. The exceptional case had a repeat FNAC, which yielded a malignant result, and histopathological examination of the partial mastectomy specimen of this case of invasive breast cancer was reported as Nottingham grade 3 malignancy. Two illustrative PBDA cases are demonstrated in Figure 1.

Close

Figure 1:

(a) On ×40 magnification, a liquid-based cytology slide demonstrates a proliferative breast lesion with mild anisonucleosis and nuclear overlapping. The slide consisted of several groups similar to the microphotograph. Therefore, the case was reported as proliferative breast disease with atypia (PBDA) (scale bar: 75 µm). (b) The histopathological correlation is a fibroadenoma as seen on the ×4 magnification (scale bar: 750 µm). (c) On ×40 magnification of the Pap-stained slides, a proliferative epithelial cell group with nuclear overlapping and mild nuclear enlargement is seen. As it raised suspicion of malignancy, the result of the fine-needle aspiration cytology performed was reported as PBDA (scale bar: 75 µm). (d) On ×4 magnification of the excisional biopsy specimen, an invasive carcinoma with an extensive ductal component is demonstrated (scale bar: 750 µm).

Export to PPT

The mean ages of the patients in the benign and malignant groups were 42.1 and 54.8 years, respectively. In the ANOVA, older age was associated with malignant histological outcome (P < 0.001). The test results and the age pyramid showing two different outcomes are demonstrated in Figure 2. The majority (n = 22: 68.8%) of 32 cases aged ≥50 years were included in the malignant group.

Close

Figure 2:

(a) The analysis of variance test based on age shows a significant difference between the malignant and benign histological outcomes. (b) The age pyramid, showing two different outcomes, demonstrates the age-based population in 5-year increments.

Export to PPT

DISCUSSION

This study cohort includes cases with PBDA diagnosed based on FNAC results. Comparable to some literature reports, only 3.4% of 3125 breast FNAC samples received the diagnosis of PBDA.^[5,12,30,31] The incidence of PBDA was much higher (23–25%) in the data provided by Masood et al.^[4,8] In our study, the ROM was 44% among cases with PBDA. The ROM reported in the literature varies.^{[4,5,8,12,28]} The first studies conducted by Masood et al. found atypical ductal hyperplasia in more than 90% of their cases.^[4,8] Each study with a limited number of cases with PBDA found one malignant outcome in their cytohistological correlation of 3 and 11 cases.^[5,28] As the largest series up to date, the study of Zhao et al. had a 36.5% ROM in PBDA.^[12]

FNAC reports of the 30 cases that turned out to be malignant on histopathological examination were reviewed. Five of these cases were suspicious for malignancy. Regarding the other cases, the most possible explanations for the results indicative of PBDA were hypocellularity, the poor quality of the FNAC sample, and the presence of low nuclear-graded carcinomas. As an example of hypocellular aspirates, two of the cases had undergone repeat FNACs arising from the suspicion of malignancy. To illustrate the impact of poor quality and low cellularity of the samples taken on the FNAC results, FNAC procedures were repeated in two of the suspect cases to confirm their malignant potential. One of these cases was diagnosed as Nottingham nuclear-grade 3 invasive breast carcinoma in the histopathological examination of the partial mastectomy specimen. Other than this instance, all invasive and microinvasive carcinomas (38.2%) were Nottingham nuclear-grade 1 or 2.

In our study, FA constituted the most prevalent diagnosis among PBDA results. FA represents the most common diagnosis within the atypical categories.^{[1,12-14,30,31]} FNAC of fibroepithelial lesions has always been an issue for the interpreter of the cytology slides. FA typically manifests with distinctive characteristics, including a pattern of large ductal epithelial tissue fragments with myoepithelial cells, stromal fragments, and a background with a multitude of bare nuclei, as observed in FNAC.^[14] Inadequate FNAC samples and rarely seen cytological features in FA that are suggestive of a carcinoma, such as high cellularity, dispersal of intact single cells, and varying degrees of nuclear atypia, can cause misdiagnosis. The utilization of the triple test in such cases would serve to prevent misdiagnosis and mistreatment.

The results of the ANOVA demonstrated a correlation between advanced age and a malignant turn-out. This finding is consistent with the observations of Zhao et al.^[12] Based on the findings of our study, a hypothesis regarding age can be formulated. In patients aged ≥50 years, FNAC results indicating PBDA should alert the clinician.

The result of PBDA was introduced to the field of cytopathology in the early 90s by Masood et al.^[4] The reporting system was welcomed and developed further. The modified Masood’s scoring index had been a keystone for breast cytopathology until the arrival of the Yokohama system.^[1,4-6,8-16] Studies comparing these two systems could not arrive at a conclusion. ^[6,7] The Yokohama system is based on a multidisciplinary principle, and its reproducibility has been proven.^[1,13-26] The categories of atypical and suspicious of malignancy in the Yokohama system have taken the place of PBDA.^[1,4-6,8-16] PBDA constitutes a segment of the atypical and the majority of the suspicious categories of malignancy included in the Yokohama system for reporting breast cytopathology.

The main limitation of this study is the rarity of centers that use breast FNAC. Using smaller needles, CNB became the number one choice with few complications. The second limitation of this study is the outdated use of PBDA. Due to the widespread usage of the Yokohama system, it is hard to compare the results of this study with those of an up-to-date research.

SUMMARY

The histological correlation of 68 “proliferative breast disease with atypia” results was performed. This is one of the largest datasets presenting the results of “proliferative breast disease with atypia.” This study provides a comprehensive overview of these results to the cytology community. According to the authors, cytological diagnosis of a PBDA can prompt the clinician to take further action. Furthermore, the clinician needs to be more alert in cases aged ≥50 years. Given the nearly absolute positive predictive value of the triple test, this category enables cytopathologists, clinicians, and radiologists to employ the triple test method for the purpose of administering appropriate treatment.

AVAILABILITY OF DATA AND MATERIALS

The data and materials that support the findings of this study are available from the corresponding author on reasonable request.

ABBREVIATIONS

CNB: Core-needle biopsy

FA: Fibroadenoma

FNAB: Fine-needle aspiration biopsy

FNAC: Fine-needle aspiration cytology

IAC: International academy of cytology

PBDA: Proliferative breast disease with atypia

ROM: Risk of malignancy

AUTHOR CONTRIBUTIONS

IG, IO, and MAU: Decided on the concept of the study. IG: Gathered the data. IG, IO, and MAU: Interpreted the data. IG: Prepared the manuscript. IO and MAU: Revised the draft. All authors have been involved in revising it critically for important intellectual content. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work. All authors read and approved the final version of this manuscript. All the authors have read and approved the final manuscript. All authors are eligible for ICMJE authorship.