4. Measurement of chemotherapy response
Neoadjuvant chemotherapy has traditionally been used to treat locally advanced and initially inoperable breast cancer. One of the main reasons for applying a systemic therapy before rather than after curative surgery is the potential size reduction of a malignant tumor, which is thought to permit less-invasive curative surgery. In addition, clinical and pathological remission can be achieved before surgery, which can improve outcomes. Unfortunately, although in many cases a clinically meaningful remission can be achieved, not all patients benefit equally. Some tumors even increase in size despite ongoing chemotherapy, suggesting resistance. Nevertheless, neoadjuvant and adjuvant chemotherapy are still applied empirically, since no clinical tests currently exist that would allow reliably predicting the response to and benefit from a particular chemotherapy. Therefore, the effectiveness of the chemotherapy response must be assessed.
Surgical planning may be affected by tumor response patterns. Concentric shrinkage or a scattergun or honeycomb response may occur (Fig. 4); in the latter, the remaining carcinoma appears as many scattered foci over an ill-defined tumor bed . This pattern of reaction is especially challenging when planning surgical treatments since clear margins are more difficult to attain .
The effectiveness of chemotherapy can be assessed clinically or pathologically. Clinical changes from chemotherapy are measured, such as tumor shrinking, which is an indicator of a good response [24,25]. This could be evaluated after chemotherapy by physical examination, mammography, ultrasonography, computed tomography (CT) scan, or magnetic resonance imaging (MRI)  after at least 2–3 administrations of chemotherapy [25,27]. The accuracy for determining pCR in locally advanced breast cancer after neoadjuvant chemotherapy is 57% for physical exam, 74% for mammography, 79% for ultrasonography, and 93% for MRI .
Each measurement method has its advantages and disadvantages. Clinical measurement with a caliper is very easy to do but has drawbacks. Sometimes discrepancy exists between the chemotherapy response assessed by clinical examination and pathological study of the surgical specimen.
The presence of solid fibroglandular tissue and posttherapy fibrosis can cause the amount of residual illness to be overestimated during physical examination. Mammography and ultrasound had 74% and 79% accuracy for detecting postneoadjuvant pathologic tumor response, respectively, in a report of six investigations . Mammography has been demonstrated to be more sensitive than physical examination in detecting the presence of residual tumor following treatment, although it is less specific and may underestimate the degree of treatment response [29,30]. After neoadjuvant therapy, ultrasound has proven a better predictor of pathologic tumor size than mammography [28,31]. In addition, when compared to mammography and physical examination, ultrasound is the most accurate predictor of axillary lymph node response . When both modalities are negative, the combination of mammography and ultrasound appears to be the best technique for predicting complete pathologic response (80% chance) [31,33].
Breast MRI is the most sensitive imaging modality for detecting breast cancer [30,34] and the most reliable imaging modality for assessing tumor response to neoadjuvant therapy [28,35,36]. The positive predictive value (ability to accurately identify the presence of residual disease at the final pathologic examination) was high in a combined analysis of six investigations, at 93 . The negative predictive value (ability to accurately identify the absence of disease at the final pathologic examination) was only moderate (65%), lowering the overall diagnostic accuracy to 84% [28,30].
The residual tumor size measured on MRI and the pathologic tumor size determined following surgical excision are generally in good agreement. A systematic review by Lobbes et al. discovered that MRI can overestimate or underestimate the residual tumor size, with a median correlation coefficient of 0.70 (range, 0.21–0.98) .
No criteria currently exist for reporting the tumor response to neoadjuvant therapy based on imaging. The current edition of the American College of Radiology Breast Imaging Reporting and Data System lacks clear guidelines on how to submit follow-up imaging for assessing the response to therapy. Typically, the biggest dimension measurement is used to compare tumor size before and after treatment. Descriptive patterns of tumor response, such as mammographic lesion density decrease, change in internal echotexture, and concentric versus fragmented lesion shrinking with intervening normal-appearing tissue, may also be beneficial .
In clinical practice, fluorodeoxyglucose positron emission tomography (FDG PET) imaging is the most commonly used molecular imaging agent for imaging tumor glycolytic metabolism with PET. FDG PET imaging can be used for optional systemic staging and restaging of patients with stage III illness, locally progressed and inflammatory breast cancer, and recurrent and/or metastatic breast cancer, according to the most recent National Comprehensive Cancer Network guidelines. It is especially useful when the results of standard staging investigations (CT or MRI with bone scan) are inconclusive .
FDG PET has been evaluated in many studies for predicting the pathologic response to neoadjuvant treatment [, , , , ]. The largest prospective multicenter analysis included 272 examinations of 104 patients with newly diagnosed large or locally advanced non-inflammatory primary breast cancer who were also enrolled in a trial comparing two preparatory chemotherapy regimens . A threshold of a 45% drop in standardized uptake value accurately identified 11 of 15 histopathologic responders following the first cycle of chemotherapy. Nonresponders had a negative predictive value of approximately 90% (34 of 38). FDG PET imaging thus appears to help predict neoadjuvant chemotherapy response and detect early nonresponders .
The effectiveness of chemotherapy on cancer cells is measured in terms of response. In 1981, the World Health Organization (WHO) developed a method for assessing tumor response bi-dimensionally, measuring the longest size of a tumor and its size perpendicular to that length . In 1999, a new tumor response assessment method known as the Response Evaluation Criteria in Solid Tumors (RECIST) was introduced to measure uni-dimensional tumors. It was updated in 2009 (RECIST 1.1). The RECIST and WHO criteria each have advantages and disadvantages. Currently, RECIST is most used because the criteria are simpler [26,44,45].
Currently, in addition to RECIST 1.1, the Positron Emission Tomography Response Criteria in Solid Tumors (PERCIST) is available [30,46]. The response to therapy is measured as a continuous variable and expressed as a percentage difference in the SUL peak (or sum of lesion SULs) between pre- and posttreatment scans. In simple terms, a full metabolic response is the visible removal of all metabolically active tumor cells. A partial response is defined as a decrease in SUL peak of more than 30% and 0.8 units between the most intense lesion before therapy and the most intense lesion after treatment, which does not have to be the same lesion. Progressive illness is defined as an increase in SUL peak of more than 30% and new lesions of more than 0.8 units, if verified. Another indicator of advancement is a 75% rise in total lesion glycolysis .
Based on RECIST, responses are classified as a complete response, partial response, stable disease, or progressive disease. Complete response is the loss of all tumor masses. Partial response is a tumor that becomes at least 30% smaller than the longest diameter of the tumor. Stable disease is where the tumor size decrease is not enough for a partial response, but the tumor does not increase in size and become a progressive disease. Progressive disease is where a tumor increases by at least 20% of the longest diameter [24,45].
Pathological responses are more meaningful and reliable markers of life expectancy than clinical responses. However, pathological evaluation is more difficult because histopathological tissue assessment is conducted using core biopsy or during surgery. Pathological responses after chemotherapy induction in breast cancer are predictors of DFS and overall survival. Several classifications have been recommended, such as the Miller and Payne classification (Table 4), which is based on cell loss after more reliable therapy [, , ]. Based on this classification, pathological responses are divided into five levels based on the degree of death and cell damage .
In the Miller–Payne system, pathologic response is divided into five grades based on comparison of tumor cellularity between pre–neoadjuvant therapy core biopsy specimens and definitive surgical specimens. A grade of 1 indicates no change or some alteration in individual malignant cells but no reduction in overall cellularity (pathologic nonresponse); a grade of 2, minor loss of tumor cells but still high overall cellularity of up to 30% (pathologic partial response); a grade of 3, an estimated reduction in tumor cells of between 30% and 90% (pathologic partial response); a grade of 4, a marked disappearance of tumor cells such that only small clusters or widely dispersed individual cells remain, with more than 90% loss of tumor cells (almost pathologic complete response); and a grade of 5, no malignant cells identifiable in slices from the site of the tumor and only vascular fibroelastic stroma remaining, often containing macrophages, but ductal carcinoma in situ may be present (pathologic complete response) .
Clinical responses are often inconsistent with pathological responses, especially since the first stage of DNA fragmentation is difficult to evaluate with certainty. Therefore, the evaluation of specimens obtained from mastectomy is the gold standard for determining response to therapy .