Breast cancer in northern Peru: molecular subtypes and HER2 low

TL;DR

This study analyzed breast cancer subtypes in northern Peru and found that luminal B is the most common, with a notable HER2 low subgroup.

Contribution

Identified the HER2 low subgroup in breast cancer patients in northern Peru and highlighted the shift in subtype prevalence.

Findings

Luminal B subtype was most prevalent (45.6%) in northern Peruvian breast cancer patients.

HER2 low was identified in 25.1% of patients previously classified as HER2 negative.

The study emphasizes the need for targeted therapies for HER2 low patients in health policies.

Abstract

This study aimed to understand the immunohistochemical profile of breast cancer and to identify the HER2 low subgroup in the northern macro-region of Peru. A cross-sectional study was conducted in 1176 patients from the Regional Institute of Neoplastic Diseases Northern Peru, from January 2016 to December 2023. We analyzed the data (age, histological type, grade and complementary results), with frequencies and percentages. The profile corresponded to: luminal B (45.6%); luminal A (24.7%); triple negative (18.2%); and HER2 positive non luminal (11.5%). In addition, 215 patients presented HER2 low (25.1% of those previously considered negative). This study provides evidence that the subtyping of breast cancer has changed, being luminal B the most frequent. It is essential to involve health policies to acquire targeted therapies considering HER2 low patients.

El objetivo del estudio fue conocer el perfil inmunohistoquímico del cáncer de mama e identificar el subgrupo HER2 low en la macrorregión norte del Perú. Se realizó un estudio transversal con una muestra de 1176 pacientes atendidos en el Instituto Regional de Enfermedades Neoplásicas Norte del Perú desde enero de 2016 a diciembre de 2023. Los datos recolectados (edad, tipo histológico, grado y resultados complementarios), se analizaron con frecuencias y porcentajes. El perfil correspondió a: luminal B (45,6%); luminal A (24,7%); triple negativo (18,2%); y HER2 positivo no luminal (11,5%). Además, HER2 low fueron 215 pacientes (25,1% de los considerados previamente negativos). Este estudio proporciona evidencia que la subtipificación de cáncer de mama ha cambiado, siendo luminal B más frecuente, y es esencial involucrar a políticas de salud para adquirir terapias dirigidas considerando a pacientes HER2 low.

Figures

HER2 status according to complementary studies at the Instituto Regional de Enfermedades Neoplásica del Norte, Peru 2016 to 2023.

Full text

INTRODUCTION

Breast cancer is a public health problem, being the most frequent malignant neoplasm in women worldwide (31%), and the second cause of female oncologic death in low- and middle-income countries (15%) 1. In Peru, its incidence has increased by 0.5% annually since mid-2000, and its prevalence is 135 per 100,000 inhabitants 2.

Breast cancer is characterized by its heterogeneity, both morphologically and in biological behavior, clinical course and prognosis. The current classification established by the World Health Organization (WHO) considers breast cancer as: no special type (NOS), which corresponds to 75%, and special types. However, this assessment showed gaps and limitations of clinical impact 3.

With the concept of personalized medicine, breast cancer reached a more precise diagnosis with molecular classification 4. At the beginning of the 21st century, Perou et al. carried out a study that evolved into four intrinsic molecular subtypes: luminal A, luminal B, non-luminal positive human epithelial growth factor receptor 2 (HER2) gene, and triple negative. In the latter, the basal type predominates (70-80%) which, from a biological perspective, could be considered a type of cancer by its own 5.

Due to the complexity and cost of gene profiling, we were able to simulate these results by immunohistochemical (IHC) study. The basic panel consists of: estrogen receptor (ER), progesterone receptor (PR), HER2 and Ki-67, whose criteria was las modified by the European Consensus of St. Gallen in 2013 6. This intrinsic molecular subtyping is as follows: luminal A is ER positive and/or RP positive with HER2 negative and has low Ki-67 (less than 20%); luminal B is ER positive and/or RP positive, HER2 positive or negative, and has high Ki-67 (greater than or equal to 20%); HER2 positive non luminal is ER negative, RP negative, and HER2 positive; and triple negative is ER negative, RP negative, HER2 negative 4^,^6. The WHO reports that the frequency of these subtypes is as follows: luminal A (40-60%), followed by luminal B (20-30%), triple negative (basal type) (15-20%), and the HER2 positive non luminal subtype (10-20%) 3.

In Peru, four studies have described the profile of the molecular subtypes of breast cancer, three of them were developed in Lima and one in Arequipa. The first one, carried out between 2000 and 2002 by Vallejos et al., included 1198 patients and concluded that the predominant subtype was luminal A (49.3%), followed by triple negative (21.3%), HER2 positive non luminal (16.2%), and luminal B (13.2%) 7. Subsequently, Medina, in the years 2009 to 2012, in a series of 280 participants found the following rates: luminal A (37.5%), luminal B (31.4%), HER2 positive non luminal (16.4%), and triple negative (14.6%) 8. Between 2015 to 2017, Chachaima et al., conducted a study in 259 patients and reported the following: luminal A (40.1%), luminal B (32.4%), triple negative (15.4%), and HER2 positive non luminal (11.9%) 9. Finally, Zavala et al., in 2022, analyzed hormone receptor (HR) and HER2 expression in 1943 breast cancer patients, finding the following: HR positive with HER2 negative (52.4%), HR positive with HER2 positive (18.7%), HR negative with HER2 positive (12.9%), and HR negative with HER2 negative (16.0%) 10.

HER2 is an important prognostic and predictive biomarker in breast cancer. It is currently classified as positive when it scores 3+ by immunohistochemistry, or scores 2+ with HER2 gene amplification by in situ hybridization (ISH). This represents a targeted anti-HER2 therapeutic opportunity. However, the dichotomy of this receptor (positive or negative) has been overturned with the introduction of new antibody-drug conjugates (ADCs) 11^-^13.

The IHC study of HER2 was designed to differentiate high levels of its expression (almost 2 million molecules per cell for 3+ score) from lower levels (20,000 to 500,000 molecules per cell for 0 to 2+ scores). The concept of “HER2 low” has emerged in recognition of this and of the intra-tumoral heterogeneity. 11^,^12. This was defined in the DESTINY-Breast04 study as tumors expressing an immunohistochemical score of 1+, or 2+ without gene amplification by ISH. The randomized, multicenter, open-label clinical trial involved 557 patients with unresectable or metastatic HER2 low breast cancer, who after receiving targeted therapy showed improvement in overall and progression-free survival. The Food and Drug Administration (FDA) approved Trastuzumab-deruxtecan (T-Dxd) therapy in this subgroup in August 2022 11^,^13.

The aim of this study was to describe the immunohistochemical profile of breast cancer and to identify the subgroup of HER2 low patients in the northern macroregion of Peru.

KEY MESSAGESMotivation for the study: Molecular classification of breast cancer allows the use of targeted treatments. Information on this profile in the northern macroregion of Peru is unknown. In addition, new therapies have appeared for a subgroup of patients.Main findings: In this study, the most frequent molecular subtypes were: luminal B, luminal A, triple negative and non-luminal HER2. Also, 18.3% of patients had low HER2 expression.Implications: Health policies should be aligned with scientific advances, to guarantee targeted therapies and to update the information in health manuals or protocols.

THE STUDY

We conducted a descriptive, cross-sectional, retrospective study. Medical records and complementary studies were reviewed in patients with breast cancer from January 2016 to December 2023, attended at the Instituto Regional de Enfermedades Neoplásicas Norte (IREN Norte) in Peru. The inclusion criteria were to have histological diagnosis of breast cancer, complete immunohistochemistry study, and in situ hybridization (ISH) if accurate. Data were collected in Microsoft Excel 2016 and the variables of interest were: age, histologic type, histologic grade, immunohistochemistry and in situ hybridization results.

The histopathological study was carried out on core biopsies of primary breast tumors, fixed in buffered formalin and embedded in kerosene according to laboratory standards. The histological type was considered according to WHO as NOS, lobular or others. The histological grade was evaluated according to the Elston-Ellis modification of the Bloom Richardson system. The markers in the immunohistochemical study were: RE (window confirm anti-estrogen Clone SP1), RP (window confirm anti-progesterone Clone 1E2), HER2/neu (window confirm anti-Her-2/neu PATHWAY Clone 4B5), and Ki67 (window confirm anti Ki-67 30-9 Monoclonal). RE or RP was considered positive, according to the internationally agreed proportion and intensity score with an allred score greater than or equal to three points. Regarding HER2, we followed the 2018 recommendations of the American Society of Clinical Oncology (ASCO) and the College of American Pathologists (CAP), ratified at the date of the study. This immunohistochemical interpretation yields the following results: 3+ (positive), 2+ (equivocal), 1+ (negative) and 0 (negative). Where, if “equivocal”, the receptor study was continued by in situ hybridization (dual window ISH DNA Probe Cocktail Assay) with positive or negative result according to the HER2/CEP17 ratio (chromosome 17 centromeric probe). The cut-off point of 20% was considered for Ki67 according to the St. Gallen 2013 consensus discussion (3,14,15).

For the definition of breast cancer subtype we considered: luminal A (ER positive and/or PR positive with HER2 negative and Ki-67 less than 20%); luminal B (ER positive and/or PR positive, HER2 positive or negative, and Ki-67 greater than or equal to 20%); HER2 positive non luminal (ER negative, PR negative, and HER2 positive); and triple negative (ER negative, PR negative, HER2 negative). We defined cases of “HER2 low” patients as those with immunohistochemical results 1+, or 2+ with negative ISH 4^,^6^,^13.

We analyzed the variables of interest including the immunohistochemical profile in breast cancer, as well as the “HER low” cases. Frequencies and percentages are presented in tables.

This study was approved by the ethics committee of the Instituto Regional de Enfermedades Neoplásicas del Norte. The code was: 024-2023-IREN NORTE-CIEI.

FINDINGS

Of 1202 patients with breast cancer diagnosed at IREN Norte, 26 were excluded due to incomplete immunohistochemical study, resulting in a sample of 1176 patients.

The most frequent age at diagnosis was 52 years. The distribution according to the histologic type was: no special type (NOS) in 852 (72.5%); lobular in 280 (23.8%); and others in 44 (3.7%) patients. Regarding the histological grade: 18 patients were G1 (1.5%); 723 patients, G2 (61.5%); and 435 patients G3 (37.0%) (Table 1).

Table 1. Age and histological characteristics according to molecular subtype of breast cancer in patients from the Regional Institute of Neoplastic Diseases of the North, Peru 2016 to 2023.  Luminal A Luminal B Luminal B HER2 HER2 Triple negative Totalcases%cases%cases%cases%cases%cases%Age 20-392815.65430.02212.22715.04927.218015.3 40-5914622.419429.812018.47411.311818.165255.4 60-7910332.79730.84113.0319.84313.731526.8 80 or more1448.3620.7310.326.9413.8292.5Histological type NOS18221.425529.913315.610512.317720.885272.5 Lobular8831.49132.54716.8258.92910.428023.8 Other2147.7511.4613.649.1818.2443.7Histological grade G11161.1316.715.615.6211.1181.5 G222831.522130.612016.68111.27310.172361.5 G35212.012729.26514.95212.013932.043537.0HER2: human epidermal growth factor receptor 2, NOS: no special type, G1: histologic grade 1, G2: histologic grade 2, G3: histologic grade 3

Regarding the molecular profile, the most frequent subtype was luminal B in 537 patients (45.6%), of which 186 expressed HER2, followed by luminal A with 291 (24.7%), triple negative in 214 (18.2%) and finally HER2 positive non-luminal in 134 (11.5%). When evaluating the distribution by year, we found that in 2016, 50.0% were luminal A cases, decreasing to less than half in the year 2023 with 21.0%. The opposite occurred with the luminal B subtype, which in 2016 was 21.4%, and in 2023 represented 54.8%. The subtypes: triple negative and HER2 positive non luminal, have shown little variation over the years, with a slight increase in the latter from 8.9% (2016) to 10.8% (2023) (Table 2).

Table 2. Molecular subtyping of breast cancer by year at the Regional Institute of Neoplastic Diseases of the North, Peru 2016 to 2023.YearLuminal A Luminal B Luminal B HER2 HER2 Triple negative Total by yearCases%Cases%Cases%Cases%Cases%Cases%20162850.058.9712.558.91119.6564.820173337.91416.11719.5910.31416.1877.420185032.92717.82315.12617.12617.115213.020195229.95229.92514.4169.22916.717414.820202616.15332.92616.1138.14326.716113.720212314.27043.22012.32213.62716.716213.820223820.16433.92714.32211.63820.118916.020234121.06633.84121.02110.82613.319516.5Total 29124.735129.818615.813411.521418.21,176100.0HER2: human epidermal growth factor receptor 2.

We obtained the following HER2 expression and interpretation results, the most frequent mode was negative (score 0) in 642 patients, negative (1+) in 105 patients, equivocal/doubtful (2+) in 167 patients, and positive (3+) in 262 patients. After the ISH study in equivocal cases, patients expressed dichotomous results: positive (319 patients) and negative (857 patients). The “HER2 low” case definition was applied in the latter group, with 215 patients (18.3% of the total sample and 25.1% of the negative subgroup) meeting these criteria (Figure 1).

Figure 1HER2 status according to complementary studies at the Instituto Regional de Enfermedades Neoplásica del Norte, Peru 2016 to 2023.

Of the latter group, 131 patients had intermediate histologic grade (G2) and 163 patients were histologic type NOS. Regarding the molecular subtype of the HER2 low patients, 78 patients were luminal A, 107 luminal B, and 30 triple negative (Table 3).

Table 3. Characteristics of patients with “HER2 low” breast cancer at the Regional Institute of Neoplastic Diseases of the North, Peru 2016 to 2023. Luminal A Luminal B Triple negative Total Cases%Cases%Cases%Cases%Age 20-39925.71954.3720.03516.3 40-594235.96152.11412.011754.4 60-792341.12646.4712.55626.0 80 or more457.1114.3228.673.3Histological type NOS5332.58753.42314.116375.8 Lobular2043.52043.5613.04621.4 Other583.300.0116.762.8Histological grade G1133.3266.700.031.4 G25844.36247.3118.413161.0 G31923.54353.11923.58137.6 Total 7836.210749.83014.0215100.0NOS: no special type, G1: histologic grade 1, G2: histologic grade 2, G3: histologic grade 3.

DISCUSSION

In this study we analyzed the distribution of molecular subtypes of breast cancer, as well as data with prognostic value and those considered “HER2 low”. The most affected population group was those between 40 and 59 years of age. The American Cancer Society recommends that people aged 45 to 54 years should undergo annual mammography screening. In addition, the most frequent breast cancer in young women aged 20 to 40 years was triple negative, which is known to be more aggressive. Breast cancer in young women represents a significant burden for low- and middle-income countries, with more than 20% of cases occurring in women under 45 years of age 1^,^2^,^15.

Similar to our study, the highest proportion of high-grade breast cancer (G3) corresponded to the triple negative and HER2 non luminal subtypes in India, United States, Iran, Egypt, Morocco, Japan and Korea. On the other hand, most low-grade tumors (G1) were luminal A type 3^,^9^,^16. The most frequent histological grade was G2 (61.5%), followed by G3 (37.0%) and G1 (1.5%). The same order of frequencies was reported by Firdaus et al., (G2 - 62.1%, G3 - 26.8% and G1 - 9.1%) 17.

The molecular profile in the 1176 patients from our study differs from the international and national literature. According to WHO, the most frequent subtype is luminal A with 40 to 60%, and the national studies reviewed coincide with this information. However, in our population it was in second place with only 24.7%. On the other hand, luminal B breast cancer was the most frequent subtype with 45.6% compared to 20.0% reported by the international literature and up to 32.0% in national studies 7^,^8^,^9.

In addition, the analysis by year showed that the frequency varied. From 2016 to 2023 there was an inversion of percentages of luminal A and luminal B breast cancer, in favor of the latter. This is a worrying situation since luminal A tumors have better prognosis, longer survival and lower recurrence among all subtypes 3^,^8. This may be due, in theory, to two circumstances, that there really is an increase in luminal subtype B, or that diagnostic tests have improved its recognition. In this article, we believe that both situations are feasible.

The first scenario could be caused by new mutations as some studies suggest that luminal B and other hormone-resistant breast cancer subtypes evolve from luminal A cancer. Estrogen being a dynamic regulator of several factors, loss of estrogen function and increased growth factor receptor (GFR) may occur during breast cancer progression to a hormone-resistant state 18.

In the second scenario, the role of the pre-analytical phase was of vital importance, standardizing the optimal processes for reliable immunohistochemical and molecular results. Thus, it became known that prolonged ischemia by late fixation decreases receptor expression (PR begins to decrease after one hour, ER after two hours, and after eight hours both expressions become completely negative) and that the fixation time for optimal receptor expression is a minimum of six to eight hours, while the maximum fixation time without a change in ER, PR, HER2 and ki67 expression is 72 hours. This led to the development of international guidelines and recommendations that regulated the time of ischemia, fixation, processing and reading, in order to limit the pre-analytical variables achieving a more accurate immunohistochemistry 19.

On the other hand, we evidenced a similar frequency of the most aggressive subtypes, triple negative (18.2%) and HER2 positive non luminal (11.5%), compared to the most current national study with 16.0% for triple negative (negative HR with HER2 negative) and 12.9% for HER2 positive non luminal (negative HR with HER2 positive) 10.

When HER2 status was assessed, we found that 215 patients met the criteria for the “low” subgroup. However, there is no national data to compare this finding. Internationally, it is reported that almost half of HER2-negative breast cancers show some degree of expression (“HER2 low”), but in our study it represented only 25.1% 12^,^13^,^20^,^21.

The molecular classification has not changed and HER2 low cannot be considered a molecular subtype, but it is important to recognize these patients since in the DESTINY-Breast04 clinical trial, trastuzumab deruxtecan (T-DXd) demonstrated that even fewer HER2 receptors on cancer cells may be sufficient for clinical benefit 20^,^21. Evaluation of this subgroup, especially in those triple-negative patients, is vital as it will allow targeted therapies, new clinical trials and the development of more accurate techniques to assess HER2 status 22.

Our study had some limitations such as the lack of national or regional statistics on HER2 low patients for comparison of results, the absence of data in the medica records on the current status of the patients, making it difficult to analyze the survival of a large part of the sample, and the absence of genetic tests that could enrich the results.

The subtyping profile of breast cancer in patients diagnosed at the Northern Regional Institute of Neoplastic Diseases during 2016 to 2023 differs from previous studies, showing a notable increase in luminal B subtype breast cancer at the expense of a decrease in luminal A subtype. Likewise, we found that the HER2 low subgroup represents a quarter of those patients previously considered without expression of these receptors, opening a therapeutic opportunity for them. This implies that public health policies must be involved and change according to the context. There must be a permanent contribution to research and more effective early detection programs, besides, specific treatments must be acquired, trained human resources must guarantee, and alliances for personalized medicine must be created.