Pathology of Neoadjuvant Therapeutic Response of Breast Carcinoma

Introduction

Administration of chemotherapy or endocrine therapy before definitive surgery (neoadjuvant systemic therapy [NST]) is a standard treatment for inflammatory and inoperable locally advanced breast cancer. With advances in breast imaging and improvements in tumor characterization and staging, patients who are eligible for adjuvant chemotherapy or endocrine therapy for operable early stage breast cancer are now being offered NST. Clinical trials have shown no difference in locoregional control and metastasis-free survival between patients who receive adjuvant therapy versus NST; however, NST provides a unique opportunity to assess the efficacy of therapy in vivo and renders more women eligible for breast conserving surgery (BCS). In addition, tumor response is used as a short-term end point that can be measured in months rather than years of follow-up, particularly for clinical trials and evaluation of the efficacy of new agents. Tumor response to systemic agents also serves as a powerful surrogate for long-term survival.

Predictors of Therapeutic Response

The categories of therapy response—namely, complete response (CR), partial response (PR), minimal or no response (NR), and progressive disease (PD)—are defined by the change in tumor size from pretreatment clinical or radiological measurements to posttreatment clinical, radiological, and pathological measurements ( Fig. 28.1 ). Currently, pathologically complete response (pCR) is defined as the disappearance of all invasive carcinoma in the breast and in the axillary lymph nodes after completion of NST. Most systems allow the presence of residual carcinoma in situ because this finding does not alter overall survival (OS). Even though the rate of pCR varies from study to study based on the tumor type and the therapeutic regimen used, most report that patients who achieve pCR (responders) have improved long-term, disease-free survival (DFS) than nonresponders.

Fig. 28-1, Types of response to neoadjuvant systemic therapy. (1) Minimal response to neoadjuvant systemic therapy results in a minimal decrease in tumor size, which is evidenced by a narrow zone of fibrosis around the perimeter of the tumor. (2) Partial response results in multiple scattered tumor nests and single cells throughout the tumor bed or a small contiguous residual tumor nodule. (3) Complete response leaves a fibrosclerotic area without tumor cells.

The clinical and pathological predictors consistently associated with a better response to therapy include tumor subtypes, tumor size, high proliferation index, presence of necrosis, tumor-infiltrating lymphocytes (TILs), and gene expression profile.

Tumor subtypes based on immunohistochemical (IHC) studies have shown that patients with hormone receptor–negative (HR−) tumors achieve significantly higher pCR rates than those with HR+ tumors. For example, patients with invasive lobular carcinoma and receptor-positive ductal carcinomas rarely achieve pCR, whereas patients with triple-negative tumors (negative for estrogen receptor [ER], progesterone receptor [PgR], and human epidermal growth factor receptor 2 [HER2]) achieve higher rates of pCR. One study reported 7% pCR in HR+ tumors (luminal subtypes) compared to 27% and 36% pCR for basal-like and HER2 + /ER− subtypes, respectively. Patients with HER2+ tumors who received anti-HER2 therapy as part of a neoadjuvant regimen achieve higher pCR rates (up to 60%) compared to those who did not. Furthermore, patients who received dual anti-HER2 agents in combination with chemotherapy achieved higher pCR rates than patients who received a single anti-HER2 agent. However, the greatest benefit of anti-HER2 agents in the neoadjuvant setting is seen in patients who have HR−/HER2+ tumors versus HR+/HER2+ tumors. In fact, the benefit of trastuzumab progressively decreases with increasing ER expression in the tumor.

Studies that analyzed TILs have shown that the presence of specific subtypes of lymphocytes correlates with response to therapy. A high level of CD8+ cells and a high ratio of CD8+/T regulatory cells (Treg/FOXp3+) in triple-negative breast cancers (TNBCs) correlates with higher rates of pCR after neoadjuvant chemotherapy. A higher ratio of CD8+/FOXp3+ before and after neoadjuvant therapy is correlated with response to therapy in both TNBC and HR+ tumors. Studies examining immune-modulating genes in TNBC and HER2+ tumors have shown expression of certain immune genes/metagenes predictive of pCR in the breast.

In HR+/HER2− tumors, the proliferation rate assessed by Ki-67 expression can be used to tailor neoadjuvant endocrine therapy or chemotherapy. In HR+/HER2− tumors, best response to chemotherapy is associated with high Ki-67 expression, which has been shown by multivariable models such as Magee Equation 3 (ME3) and immunohistochemical 4 (IHC4).

Studies have used gene expression profiling of tumors to predict response to therapy. A retrospective study by Rodriguez et al revealed that defective DNA repair gene expression signature status in sporadic TNBC differentiates tumors that are sensitive to anthracyclines and resistant to taxane-based chemotherapy. Genotyping of HER2+ tumors from patients who received anti-HER2 therapy in a neoadjuvant setting has shown that the presence of activating mutations in the gene encoding the phosphatidylinositol 3-kinase catalytic subunit (PIK3CA) is associated with lower pCR. Similarly, gene expression profiles in pretreatment biopsies from locally advanced HR+ tumors in postmenopausal patients have been validated to predict response to selective ER antagonists.

Evaluation of Therapeutic Response

The response of cancers to systemic agents can be assessed during treatment by clinical examination and by imaging studies ( Fig. 28.2 ). Physical examination and routine imaging studies such as mammography and ultrasonography to measure residual tumor have high interobserver variability and are prone to error due to frequent overestimation or underestimation. Pathological examination of the posttreatment specimen is the gold standard and currently is the only method that can accurately determine the presence and extent of residual disease.

Fig. 28-2, Response to chemotherapy monitored by serial magnetic resonance imaging studies. ( A ) Locally advanced right breast carcinoma before therapy demonstrates an 11 × 6-cm mass with enhancement. ( B ) Two months later, the mass measures 8 × 6 cm. ( C ) At the end of chemotherapy, the mass is 6 × 3 cm without significant residual enhancement.

Clinical Evaluation

Tumors with little or no response to NST remain palpable after therapy, but tumors with pronounced response to treatment may no longer be palpable. The change in palpability is related to softening of the tumor stroma (i.e., the quality of the desmoplastic reaction and decrease in cellularity and vascularity). Typically, patients who have a pronounced clinical response to NST have better long-term survival.

Mammography and ultrasonography, play a critical role in the initial detection of breast cancer, but they provide little quantitative information on response to therapy beyond size calculations. Recent breast imaging modalities such as dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) and ¹⁸ F-fluorodeoxyglucose (FDG) positron-emission tomography (PET) have shown promising results in the evaluation of treatment response. Measurement of tumor vascularity and functional volume by DCE-MRI before and during treatment can predict response to therapy. PET can assess blood flow and metabolism using ¹⁵ O-water and ¹⁸ F-FDG, both before and during therapy, to predict response to chemotherapy. Because ¹⁵ O-water PET poses a logistical challenge caused by the short half-life of ¹⁵ O, studies have used a combination of methods including DCE–computed tomography (CT) or DCE-MRI in conjunction with FDG PET to measure tumor perfusion and vascularity. The combined FDG PET/CT approach has been shown to be superior to MRI in predicting pCR earlier in the course of treatment in some studies. Unfortunately, limited availability of combined PET/CT or PET/MRI systems prevents routine use of these imaging modalities to evaluate tumor response.

Key Clinical Features

CNB , Core needle biopsy; ER , estrogen receptor; FNA , fine-needle aspiration; HER2 , human epidermal growth factor receptor 2; NST , neoadjuvant systemic therapy; PgR , progesterone receptor.

Neoadjuvant Therapy

■
NST refers to the treatment of patients with systemic agents before definitive surgical removal of a tumor.
■
Pathological assessment of response to therapy provides valuable information for further management and prognosis for individual patients.
■
A definitive diagnosis of invasive carcinoma must be made by CNB.
■
There must be sufficient tissue to evaluate tumor markers (ER, PgR, HER2, Ki-67).
■
A standard protocol for evaluating lymph nodes should be followed:
- ■
  Clinically positive nodes should be documented with FNA or core biopsy before therapy.
- ■
  Clinically negative nodes should be confirmed by imaging studies before therapy.
■
A clip must be placed at the time of CNB to mark the site of the tumor, both in the breast and in the lymph node.
■
It is essential that the entire tumor bed be excised with a rim of normal breast tissue.

A clip should be placed at the time of core needle biopsy (CNB) or during the first few cycles of NST, ideally before the tumor is too soft or unrecognizable on imaging studies. If a clip is not placed, it may not be possible to reliably identify the tumor bed. Although calcifications when associated with a carcinoma usually remain after treatment, they are infrequent and unreliable. Ideally, candidates who become eligible for BCS after NST should undergo a form of localized excision such as seed or wire.

Pathological Evaluation

Pretreatment Tumor Evaluation

As mentioned earlier, a core biopsy of the pretreatment tumor should have an adequate representation of invasive carcinoma where testing and interpretation of biomarker studies can be completed. Ideally, the core biopsy report should include the following information: tumor type, tumor grade, presence of necrosis, lymphatic and vascular invasion (LVI), and associated stromal TIL if present. Providing in the pathology report the number of cores that contain invasive tumor can help clinicians determine the adequacy of the tumor sample and the reliability of the ER, PgR, and HER2 test results. Although not a standard requirement at the time of this writing, many pathology laboratories routinely perform a Ki-67 IHC to report the proliferation index in all invasive breast carcinomas. This can be an important adjunct if NST is planned because a Nottingham grade may not be accurate. Currently, pathologists do not quantitate or characterize TILs if present, but this may be required if a patient is enrolled in a clinical trial. Similarly, some trials may require testing for PDL1 in a core biopsy of TNBC to determine patients’ eligibility for checkpoint inhibitors.

Pretreatment Lymph Node Evaluation

Determination of nodal status by clinical examination alone may not be accurate. Fine-needle aspiration (FNA) or CNB of a clinically palpable or radiologically suspicious node is the best method to establish node positivity. A biopsy clip or tattoo mark in the node enables the surgeon to remove the node and allows the pathologist to evaluate the node after completion of systemic therapy.

It can be challenging to perform sentinel lymph node biopsy (SLNB) after NST due to fibrosis in the nodes. Several studies have reported a 10% to 20% false-negative rate for finding sentinel lymph nodes (SLNs) after NST. For this reason, some consider SLNB before NST in patients who have clinically negative nodes. However, SLNB before completion of NST will preclude evaluation of treatment response if SLN reveals metastases. Though a multicenter neoadjuvant chemotherapy trial (the Alliance trial) reported a 12.6% false-negative rate of SLNB in patients with clinically positive nodes (cN1), a majority were able to retrieve at least two SLNs during surgery. It has been shown that placing a biopsy clip in a positive node prior to NST and removing the clipped node (which may or may not be the sentinel node) can reduce the false-negative rate.

Historically, patients who have had a biopsy-documented positive lymph node before NST have undergone completion axillary lymph node dissection (ALND) at the time of resection of the primary tumor. However, 22% to 42% of patients with proven positive nodes achieve pCR in the nodes. Because this subgroup of patients can be spared from the morbidity of ALND, many centers currently perform SLNB and intraoperative assessment of the SLN to determine whether to proceed with ALND or not.

Posttreatment Tumor Evaluation

Gross Examination

Identification of tumor bed

To assess patient response to treatment, the key is to identify the original tumor site (tumor bed). It is easy to identify residual tumor if there has been no or minimal response as the mass remains firm and palpable after treatment. Tumors that demonstrate complete or near-complete resolution on clinical examination and on imaging studies are unlikely to be visible on gross examination ( Figs. 28.3 and 28.4 ). If the tumor bed is not apparent or the biopsy clip is not identified on initial gross examination, it is prudent to obtain a specimen radiograph to locate the biopsy clip. Lacking these, one must rely on the microscopic detection of the tumor bed in partial mastectomy specimens to ensure the prior tumor site was removed.

Fig. 28-3, Pre- and posttreatment images of breast carcinoma. ( A ) Pretreatment mammographic appearance of an invasive breast cancer represented by a large area of abnormal density. ( B ) Mammographic appearance of the same tumor after chemotherapy demonstrates a marked reduction in the area of abnormality. ( C ) Posttreatment mastectomy specimen demonstrates a vague grossly visible fibrotic tumor bed without grossly identifiable residual tumor.

Fig. 28-4, Pre- and posttreatment contrast-enhanced magnetic resonance imaging (MRI) scans. ( A ) Pretreatment maximum intensity projection (MIP) image of a patient presented with a 1.5-cm triple-negative, node-positive breast carcinoma correlating with a 1.2-cm area of intensity on MRI. ( B ) The patient received four cycles of chemotherapy. Posttherapy MIP image shows complete resolution of the abnormality. ( C ) Posttreatment partial mastectomy was serially sectioned to reveal no grossly visible tumor bed, but sections from the clip area showed changes of the tumor bed (not shown).

For total mastectomy specimens, a detailed description of the pretreatment tumor location (i.e., quadrant, distance from nipple) or sutures placed by the surgeon is extremely helpful in finding the tumor bed. In our experience, specimen radiograph of the mastectomy specimen is invaluable to locate the tumor bed if the biopsy clip or the tumor bed is not apparent on initial sectioning.

The tumor bed, when visible, has the appearance of a nondescript area of irregular rubbery fibrous tissue (see Figs. 28.4C and 28.5 ). Residual tumor may be recognized as tan nodules within the tumor bed ( Fig. 28.6 ). It is important to document the size of the grossly visible tumor bed and residual tumor in two dimensions. In rare instances, smaller tumor with complete response may not leave any residual grossly visible fibrous scar ( Fig. 28.7 ). In those cases, the tissue around the biopsy clip should be submitted for microscopic examination.

Fig. 28-5, Posttreatment mastectomy specimen with a vague fibrotic tumor bed without grossly identifiable residual tumor. On pathological examination, small foci of residual tumor cells were identified (not shown).

Fig. 28-6, Partial mastectomy specimen with small tan nodules of residual tumor in an irregular fibrotic tumor bed.

Fig. 28-7, Total mastectomy specimen with a coiled clip embedded within hydrogel (arrow) with complete resolution of the tumor.

Evaluation of margins

Larger tumors with pronounced response to treatment are difficult to localize both on imaging studies and clinically by the surgeon. Therefore, both partial and total mastectomy specimens should be inked to assess margins in the event that residual tumor is found on microscopic examination.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here