Keywords: Neoadjuvant therapy - Breast neoplasms - Pathology - Surgical - Neoplasm - Residual.
Descritores: Terapia neoadjuvante - Neoplasias mamárias - Patologia Cirúrgica - Neoplasia residual.
INTRODUCTION
Breast cancer is the most common malignancy among women both in developed and in developing
countries[1 ]. In Brazil, it has been estimated that there will be 66,280 new cases of breast
cancer for each year of the triennium 2020-2022, which corresponds to an estimated
risk of 61.61 new cases per 100,000 women[2 ].
Breast cancer treatment strategies are defined according to clinical and pathological
findings and predictive and prognostic factors such as staging and molecular subtype.
In practice, breast cancer subtypes are identified by means of immunohistochemistry
and are classified as luminal, amplified HER2 or triple negative (TN). TNM is the
international system that is used to evaluate the extent of neoplasia. In the latest
(eighth) edition of the TNM system, published by the American Joint Committee on Cancer
(AJCC) in 2018, pathological prognostic factors were incorporated[3 ].
Neoadjuvant treatment has been used for many years for patients with locally advanced
tumors, with the aims of enabling surgery in inoperable cases and increasing the rate
of conservative surgical procedures[4 ]. Recently, indications for neoadjuvant treatment have been expanded so that pathological
responses can be evaluated. Diversified therapeutic approaches may thus be indicated
in accordance with each residual disease profile[5 ].
The pathological response or extent of residual disease in the surgical specimen correlates
inversely with prognosis and survival. Residual tumor load is a predictor of distant
recurrence-free survival, such that cases with minimal residual disease and complete
pathological response (CPR) have a better prognosis[6 ]. Different classifications of pathological response have been used by different
authors over the years. However, it is known that CPR, defined as absence of invasive
carcinoma in the breast and axilla, confers significantly increased disease-free and
overall survival[7 ]
[8 ], especially in cases of negative receptor tumors (HER2-positive and TN)[5 ]. Thus, CPR has frequently been used in chemotherapy studies as an intermediate outcome
measuring the efficacy of treatment because it can be rapidly evaluated and reproduced.
This is a real-life study, in which the objectives are to analyze and correlate CPR
with disease-free survival and overall survival among patients with breast cancer
of different subtypes undergoing neoadjuvant chemotherapy in a private institution
in a developing country, offering treatments in accordance with international guidelines.
Periodic analysis of results obtained in an institution not only has relevance as
a management tool, thus helping to ensure the best clinical practices, but also contributes
to validation of results obtained in clinical trials.
MATERIALS AND METHODS
Methodology
This is a prospective observational cohort study on patients diagnosed with breast
cancer (men and women) who were treated with neoadjuvant chemotherapy between 2012
and 2018 in the six units of Americas Oncology, a private institution in the state
of Rio de Janeiro.
Patients were included through a prospective search, using the OpenClinica system,
Enterprise edition, for new cases of breast cancer patients who received neoadjuvant
chemotherapy. Data were collected by consulting physical and electronic medical records.
This study was approved by the Research Ethics Committee. Written informed consent
was signed by all participants.
The following patients were excluded: those who discontinued the initially planned
treatment without justifiable cause (which could be due to progression or adverse
events); those who were lost from the followup at the institution during neoadjuvant
chemotherapy; and those who were already in stage IV at diagnosis.
Histopathological and immunohistochemical analyses were performed in different local
laboratories. Immunohistochemistry was evaluated using estrogen receptors (ER) and
progesterone receptors (PR), HER2 and Ki67.
Clinical variables such as age, sex, stage and treatment protocols were collected.
Age was evaluated as age groups. For clinical and pathological staging, the TNM system
of the American Joint Committee on Cancer (AJCC), eighth edition, published in 2018,
was used. The chemotherapy regimens used for neoadjuvant treatments were as follows.
For patients who were HER2-positive: taxane + trastuzumab; taxane + anthracycline
+ trastuzumab; or associations containing double blockade with trastuzumab and pertuzumab.
For patients presenting luminal and TN disease: densedose regimens (dose-dense AC
followed by dosedense paclitaxel or dose-dense AC followed by weekly paclitaxel);
taxane alone; anthracycline alone; taxane + anthracycline; and others such as platinum.
The choice of surgical procedure (conservative or radical) was at the discretion of
the mastologist and the patient. Radiotherapy and hormone therapy were performed on
patients with indications for these, in accordance with international recommendations.
The pathological variables evaluated were histopathological subtypes and CPR. The
following subgroups were identified by means of immunohistochemistry and were defined
as: luminal A (ER-positive, PR-positive, HER2-negative and Ki67 up to 14%); luminal
B/HER2-negative (ER-positive, PR-positive, HER2-negative and Ki67 ≥ 14%); luminal
B/HER2-positive (ER-positive, PR-positive, HER2-positive and Ki67 ≥ 14%); HER2-enriched
(HER2e) (HER2-positive, ER-positive and PR-negative); and triple negative (TN) (ER-negative,
PR-negative and HER2-negative)[9 ]. Cerb2 scores were identified as 0, 1+, 2+ or 3+, and fluorescence in situ hybridization (FISH) was performed for 2+ results. Those whose results were presented
as 3+ and amplified FISH (2+ cases) were included as HER2-positive. To grade Ki67,
we used the 2013 St. Gallen Consensus, which considers that values below 14% are low
or negative[9 ]
[10 ].
The Food and Drug Administration (FDA) has defined CPR as the absence of residual
invasive neoplasia in breast and lymph node specimens after neoadjuvant chemotherapy,
while allowing the presence of residual noninvasive disease, including carcinoma in situ (ypT0/Tis ypN0, in the AJCC 8th edition)[11 ].
Statistical analysis
The results from this study were exploratory and descriptive. Overall survival was
estimated using the Kaplan-Meier method and was defined as the interval between the
date of diagnosis and death. Disease-free survival (DFS) was defined as the time interval
between the date of diagnosis and recurrence of local or distant disease. For patients
included in this study who remained alive or were lost from the follow-up, the data
were censored at the time of the last contact. P ≥ 0.05 was considered significant.
Multivariate analyses were performed between the clinical-pathological variables and
the outcomes. The statistical analysis was done using the SPSS statistical software,
version 17, IBM.
The primary objective of this study was to analyze the complete pathological response
(CPR) of patients who underwent neoadjuvant chemotherapy in a private institution
in the state of Rio de Janeiro. As secondary endpoints, we evaluated the disease-free
survival and overall survival of these patients and correlated these with clinical-pathological
variables.
RESULTS
We evaluated 198 patients, all female, with a median follow-up of 35 months. They
had ages ranging from 26 to 78 years, with a median of 48 years. Twelve percent (12.1%)
corresponded to luminal subtype A; 38.9% to luminal B/HER2-negative; 13.6% to luminal
B/HER2-positive; and 10.6% to HER2-enriched. Nine patients (18.8%) underwent FISH
for diagnostic definition. Twenty-four percent (23.7%) were of the TN subtype (as
shown in [Table 1 ]). Regarding clinical staging (cTNM grouped), stage I accounted for only 2.5% of
the cases, while stage III accounted for 43.9%. Stage II was the one most frequently
present, in 53.5% of the patients in the study.
Table 1
General characteristics of patients and treatment
n
%
Age
20-49
112
56.6%
50-59
49
24.7%
60-69
30
15.2%
≥70
7
3.5%
Receptor status subgroups
Luminal A
24
12.1%
Luminal B/HER-2-negative
77
38.9%
Luminal B/HER-2-positive
27
13.6%
HER-2e
21
10.6%
Triple negative
47
23.7%
Not classified
2
1.0%
Type of neoadjuvant chemotherapy
Anthracycline + taxane
114
57.9%
Anthracycline alone
8
4.1%
Taxane alone
2
1.0%
Dense dose
28
14.2%
Platinum
3
1.5%
Therapy directed to HER-2
Trastuzumab + pertuzumab + chemotherapy
25
12.7%
Trastuzumab alone + chemotherapy
17
8.6%
Type of breast surgery
Conservative
54
27.8%
Radical
140
72.2%
Type of axillary surgery
Sentinel lymph node biopsy
62
32%
Axillary emptying
127
65.5%
Unspecified
5
2.5%
Adjuvant radiotherapy
Yes
166
85.6%
No
19
9.8%
Unspecified
9
4.6%
Adjuvant hormone therapy
Yes
128
64.6%
No
59
29.8%
Unspecified
11
5.6%
Regarding the chemotherapy used, regimens containing taxane and anthracycline were
the ones most used, followed by dense dose. The most commonly used antiHER2 therapy
consisted of double blockade containing trastuzumab and pertuzumab. Conservative surgery
occurred in the cases of 27.8% of the patients, while 72.2% underwent radical surgery.
Regarding the axillary lymph node evaluation, 32% only underwent sentinel lymph node
excision, while 65.5% underwent axillary emptying. Adjuvant radiotherapy was performed
in 85.6%, and hormone therapy in 64.6% of the cases ([Table 1 ]).
Among the patients evaluated, four did not undergo surgery (three due to disease progression
during neoadjuvant chemotherapy and one died without known cause), which made it impossible
to evaluate their pathological response. CPR was achieved in 12.5% of luminal A cases;
19.5% of luminal B/HER2negative cases; 38.5% of luminal B/HER2-positive cases; 65%
of HER2-enriched cases; and 37.8% of TN cases. There was a significant correlation
between CPR and histopathological subtypes (p < 0.001; as shown in [Table 2 ]). Among the patients in stage I (total of five), only one reached CPR. Among the
105 patients in stage II, 34 (58.6%) achieved CPR; and among the 84 in stage III,
23 (39.7%) achieved CPR.
Table 2
Evaluation of complete pathological response (CPR) according to subtypes
Subtypes
CPR yes
CPR no
Total
Luminal A
3 (12.5%)
21 (87.5%)
24 (100%)
Luminal B/HER2- negative
15 (19.5%)
62 (80.5%)
77 (100%)
Luminal B/HER2- positive
10 (38.5%)
16 (61.5%)
26 (100%)
HER-2e
13 (65%)
7 (35%)
20 (100%)
Triple negative
17 (37.8%)
28 (62.2%)
45 (100%)
Total
58 (30.2%)
134 (69.8%)
192 (100%)
p <0.001.
Among all the patients who achieved CPR (n = 58), 91.4% (53) were under 60 years of
age (p = 0.054). Regarding Ki67, CPR was achieved in 87.3% of the cases with Ki67
that was considered positive, but without statistical significance (p = 0.23).
Disease-free survival (DFS) at the end of 24 months of follow-up, was found to be
about 90% for the patients in stage II and 80% for those in stage III (p = 0.11) ([Graph 1 ]). Regarding the subtypes, the DFS was worse for patients classified as TN and HER2enriched,
and this was statistically significant (p = 0.019), as shown in [Graph 2 ]. At the end of 36 months, the DFS for patients with CPR was 89.1%, versus 72.4%
for the others, and this was also significant (p = 0.01) ([Graph 3 ]).
Graph 1 Disease-free survival and staging (p =0.11).
Graph 2 Disease-free survival and subtypes (p =0.019).
Graph 3 Disease-free survival and CPR (p =0.01).
Overall survival was similar for stages I, II and III. At 24 months of follow-up,
it was slightly worse for stage III, but without statistical significance, with p = 0.12
([Graph 4 ]). Regarding the subtypes, the overall survival was about 97% and 94% for the luminal
B/ HER2-negative and luminal B/HER2-positive groups, respectively (p = 0.025) ([Graph 5 ]). For patients who reached CPR at the end of 36 months of follow-up, overall survival
could not be calculated, since there was no event (p = 0.08) ([Graph 6 ]).
Graph 4 Overall survival and staging (p =0.12).
Graph 5 Overall survival and subtypes (p =0.025).
Graph 6 Overall survival and CPR (p =0.08).
DISCUSSION
Neoadjuvant chemotherapy, initially used in patients with inoperable breast cancer
to improve resectability, is now commonly used for its impact on surgery, downstaging
tumours convert patients from mastectomy to breast-conservation candidates. In large
studies in the literature, the breast conservation rate with neoadjuvant chemotherapy
is around 65% compared to 49% when surgery is the initial treatment[12 ].
In our study, the results showed that conservative surgery occurred in only 27.8%
of patients and that 72.2% of patients underwent radical surgery and 65.5% underwent
axillary dissection. This contradictory result can, in part, be explained by the high
rate of patients, 43.9% in our study, who were in stage III and also by the diversity
of surgical services involved in the decision-making process, involving contradictions
inherent to each group.
CPR, especially in HER2-positive and TN tumors, has been consolidated as a prognostic
marker. Therefore, for patients with residual breast and/or axillary disease, complementary
adjuvant treatment has been recommended. Use of T-DM1 as an adjuvant after neoadjuvant
therapy for patients with residual disease in the surgical specimen has given rise
to reduction of the risk of death by 50%[13 ], which shows the benefit of neoadjuvant treatment for this type of patient. In the
case of patients with the TN subtype, the CREATE-X study showed the importance of
using capecitabine as an adjuvant after preoperative chemotherapy, with gains in disease-free
survival and overall survival[14 ]. The CPR verified in the various subtypes of our study and its correlation with
survival is in agreement with literature data[15 ].
Neoadjuvant chemotherapy protocols have been improving over the years, with higher
response rates achieved. In the context of HER2-positive cases, initial studies in
2011 already showed increased proportions of CPR and gains in survival through addition
of trastuzumab to chemotherapy[16 ]. Years later, double blockade of HER2 using pertuzumab and trastuzumab was shown
to have CPR benefit, reaching response rates of 60%[17 ]
[18 ]. In our study, most HER2-positive patients received double blockade, with CPR of
65% in HER2-enriched cases and 38% in HER2/Hormone-receptor-positive cases. These
results were similar to what has been reported in the worldwide literature. There
are few Brazilian studies on CPR data. Buzatto et al. (2017)[19 ] observed a CPR rate of 48% with use of trastuzumab alone for the HER2-enriched subtype
and 44% for HER2/Hormone-receptor-positive cases. In another Brazilian study, an even
lower rate of 33% was observed among HER2-positive patients, which can be explained
by the fact that trastuzumab was not used: at that time, this drug was not available
through the Brazilian public healthcare system[20 ]. These data reveal discrepancies in the treatment used, between different services,
especially between the private and public networks. This may have repercussions regarding
differences in survival, among women with difficulty in accessing the therapeutic
regimens recommended in international guidelines.
Minor changes were observed in neoadjuvant chemotherapy protocols for triple negative
tumors. Bayratar and Arun, in 2012[21 ], showed that there was greater benefit through use of dense-dose chemotherapy regimens
for patients with negative hormone receptors and high proliferation rates. The results
in the literature regarding use of platinum derivatives and anti-angiogenic agents
seem conflicting. The CALGB 40603 study did not show better results through use of
carboplatin and bevacizumab[22 ]. On the other hand, other trials showed that adding platinum benefited the CPR,
with values ranging from 53.2% to 58%[23 ]
[24 ]. Another agent that was tested in the neoadjuvant scenario, in TN tumors, was PARP
inhibitors, but also with controversial results[24 ]. The use of immunotherapy in TN cases has aroused great interest. In the KEYNOTE-522
study, pembrolizumab combined with chemotherapy gave rise to CPR of 64.8%[25 ]. In 2020, Mittendorf et al.[26 ] showed data on CPR rates of 58% among TN patients who used chemotherapy consisting
of nab-paclitaxel and anthracycline in combination with atezolizumab. However, it
remains unknown within the immunotherapy scenario whether CPR correlates with overall
survival. In our study, most patients used regimens containing a dense dose, and only
three used platinum. We observed high response rates, with CPR in 37.8% of the cases
of TN, which seems similar to what has been reported in the worldwide literature,
as reported in the review by Asaoka et al., in 2020[27 ], in which the CPR rate was 34.2%. In the Brazilian literature, we found lower CPR
rates (21%), again in a public institution, with low financial resources and patients
with advanced disease[28 ].
Unlike TN and HER2- positive cases, in which the CPR rate correlates with the prognosis,
luminal subtypes A and B do not show any close correlation, according to the 2012
publication by Journal of Clinical Oncology[29 ]. However, for patients with positive hormonal receptors, previous studies showed
that higher clinical response rates were found through use of dose-dense chemotherapy,
such that even conservative surgery became possible. In luminal tumors, the CPR rate
was much lower than that of the previous subtypes, ranging in the literature from
6.4% to 22%[30 ] for luminal A tumors and 11% to 28%[29 ]
[31 ] for luminal B tumors, similar to the data found in the present study (luminal A
CPR of 12.5% and Luminal B CPR of 19.5%).
This study had a median follow-up of 35 months. Disease-free survival (DFS) at the
end of 24 months of follow-up, was about 90% for patients in stage II and 80% for
those in stage III, and the overall survival was similar for stages I, II and III.
It was slightly worse for stage III, but this difference was not statistically significant.
These data differed from what had been reported the literature because it is recognized
that staging is a prognostic factor for survival. This may perhaps be explained by
the small number of patients at an early stage.
In an attempt to identify predictive factors for CPR and prognostic factors for survival,
we conducted multivariate analysis (subtypes, staging, age and Ki67). We found a significant
correlation between histopathological and CPR subtypes, as well as in relation to
DFS and overall survival. Asaoka et al., in 2020[27 ], found response rates of 52.9% for HER2enriched cases, 34.2% for TN cases and 14.7%
for luminal cases. These data are similar to what we observed in this study (65%,
37.8% and 19.5%, respectively, for HER2-enriched, TN and Luminal B/HER2-negative cases)[32 ]. Regarding staging, the literature shows that cases in initial stages correlate
with higher rates of CPR[29 ]. However, in our study, we did not find any correlation between staging and CPR
(p = 0.67). Among the patients in stage I (total of 5), only one achieved CPR. Among
the 105 patients in stage II, 34 (58.6%) achieved CPR and among the 84 in stage III,
23 (39.7%) achieved CPR. Regarding age, the median was 48 years, ranging from 26 to
78, and there was no relationship with CPR or survival. Regarding Ki67, we found that
CPR was achieved in 87.3% of the cases in which Ki67 was considered positive (p = 0.23),
with worse disease-free survival outcomes, which is consistent with data found in
a meta-analysis by Tao et al.[33 ].
Both in our study and in the study by Minckwitz et al. (2012)[29 ], CPR was associated with better prognosis for HER2-enriched and TN cases, and it
was correlated with better DFS in luminal B/HER2-negative, HER2enriched and TN cases.
Thus, our data are similar to those of the worldwide literature and reflect good access
to the therapies currently existing, which are already incorporated in the private
healthcare system of Brazilian society. It is noteworthy that despite being a prospective
study, many follow-up losses occurred through exchanges of health insurance, thereby
decreasing the number of patients evaluated and impacting on some results.
CONCLUSION
In this study, we confirmed the correlation between complete pathological response
and overall survival. Thus, it is essential that increased attention is given to indications
for neoadjuvant treatment, especially in the triple negative and HER2-positive subgroups,
for which CPR has better prognostic value. In this study, we were able to show that
even in developing countries such as Brazil, adequate treatments that are in accordance
with international guidelines can be offered. The consequence of this is that our
results are similar to those in the worldwide literature. However, it is essential
that the coverage of these therapies should be expanded to encompass the entire private
network and, especially, the public network. In this manner, equal treatment, with
similar and fair outcomes for breast cancer patients with locally advanced scenarios
can be provided.
Further studies with assessments such as this one should be encouraged, so that better
understanding of the results in countries with more deficient health structures can
be obtained, thereby improving access to the most recommended therapies worldwide.
Bibliographical Record
