TL;DR
This study reveals that breast cancer's molecular subtype, specifically HER2-overexpressing and triple-negative cancers, independently predicts a lower likelihood of receiving immediate post-mastectomy reconstruction, even after accounting for other patient and tumor characteristics. This finding suggests that the inherent biological aggressiveness of these subtypes may influence reconstructive decisions, particularly in more advanced or aggressive disease presentations.
What Problem Does This Paper Address?
Immediate post-mastectomy reconstruction has become a common option for breast cancer patients, aiming to improve quality of life and body image. However, the decision to undergo reconstruction can be influenced by various factors, including patient demographics, socioeconomic status, and tumor characteristics. A key question remained: does the specific molecular subtype of breast cancer itself play a role in whether a patient opts for immediate reconstruction? Understanding this is crucial because different subtypes have distinct prognoses and treatment strategies, and some are associated with a higher risk of local recurrence. This paper aims to fill this knowledge gap by investigating the association between breast cancer subtypes (defined by hormone receptor and HER2 status) and the incidence of immediate post-mastectomy reconstruction in the United States.
Methods
The researchers utilized a large, population-based dataset from the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) program. They analyzed data from stage I-III breast cancer patients who underwent either mastectomy alone or mastectomy followed by immediate reconstruction between 2010 and 2012.
The core of their analysis involved:
- Defining Subtypes: Breast cancers were categorized into four subtypes based on immunohistochemistry for estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2): HR+/HER2-, HR+/HER2+, HR-/HER2+ (HER2-overexpressing), and HR-/HER2- (triple-negative).
- Univariate Analysis: This initial step used chi-square tests to identify individual factors (like age, race, marital status, tumor stage, and cancer subtype) that were associated with the decision to have immediate reconstruction.
- Multivariate Logistic Regression: To isolate the independent effect of each factor, a multivariate model was employed. This allowed the researchers to control for the influence of other variables and determine which factors, including breast cancer subtype, significantly predicted the likelihood of receiving immediate reconstruction. Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) were used to select the best regression model.
- Subgroup Analysis: To explore nuances, the researchers examined whether the association between subtype and reconstruction rates differed based on tumor grade and stage (AJCC stage).
Datasets
The primary dataset used in this study was the Surveillance, Epidemiology, and End Results (SEER) database. Specifically, the SEER 18 registry database (November 2014 submission) was utilized, covering incidence data from 1973-2012. The study focused on cases diagnosed between 2010 and 2012 because HER2 status data became reliably available in SEER starting in 2010. This is a large, comprehensive dataset representing approximately 28% of the US population, making it suitable for population-level analyses. No new benchmarks were introduced; the study leverages existing, well-established national cancer registry data.
Key Insights and Ideas
- Subtype as an Independent Predictor: The most significant insight is that breast cancer subtype is not just correlated with outcomes but is an independent factor influencing the decision for immediate reconstruction, even when accounting for a wide array of other patient and clinical variables.
- Aggressive Subtypes, Less Reconstruction: The finding that HER2-overexpressing and triple-negative subtypes are associated with lower reconstruction rates is particularly interesting. This suggests that the biological aggressiveness and higher risk of local recurrence inherent to these subtypes might play a role, either through patient concerns or physician recommendations.
- Grade and Stage Matter: The subgroup analysis revealing that the subtype-reconstruction association is only significant in higher-grade and more advanced-stage cancers is a crucial nuance. This implies that for less aggressive, early-stage disease, the biological subtype might be less of a deciding factor in reconstructive choices.
- Reconstruction Type is Consistent: Interestingly, the type of reconstruction (implant vs. tissue-based) did not significantly differ across subtypes, suggesting that reconstructive techniques are more influenced by patient factors like BMI or surgical expertise rather than the cancer's molecular profile.
Key Results
- Out of 47,123 eligible patients, 32.2% received immediate post-mastectomy reconstruction.
- Reconstruction Rates by Subtype (Unadjusted):
- HR+/HER2-: 33.1%
- HR+/HER2+: 33.1%
- HR-/HER2+ (HER2-overexpressing): 29.6%
- HR-/HER2- (triple negative): 27.7%
- A chi-square test showed a statistically significant difference (P < 0.001).
- Multivariate Analysis Results: After adjusting for numerous factors, HER2-overexpressing patients were 11% less likely (OR 0.896, 95% CI 0.817–0.984, P=0.021) and triple-negative patients were 19.4% less likely (OR 0.806, 95% CI 0.751–0.866, P<0.001) to undergo immediate reconstruction compared to HR+/HER2- patients.
- Subgroup Analysis: The difference in reconstruction rates between subtypes was significant for patients with high-grade tumors (poorly or undifferentiated) and advanced stage disease (AJCC stage III) (P=0.011). However, no significant difference was observed for patients with low-grade tumors (well differentiated) and early stage disease (AJCC stage I) (P=0.286).
Limitations and Future Work
The authors acknowledge several limitations:
- Missing Data: The SEER database does not include crucial information such as margin status, neoadjuvant chemotherapy receipt, hospital name, or detailed comorbidities (e.g., obesity, diabetes, smoking status, cardiac/pulmonary disease), which could influence reconstructive decisions.
- Approximation of Subtypes: The classification of subtypes based on immunohistochemistry is an approximation of true molecular subtypes.
- Physician Bias: The potential influence of physician bias in steering patients towards or away from reconstruction could not be assessed.
Future work suggested by the authors includes:
- Investigating Underlying Reasons: Further research is needed to uncover the specific reasons behind the observed disparities in reconstruction rates among different subtypes. This could involve exploring patient concerns, physician decision-making processes, and the impact of treatment side effects (e.g., from neoadjuvant chemotherapy, which was not available in the SEER data).
- Broader Comorbidity Data: Future studies could benefit from incorporating more detailed comorbidity data to provide a more comprehensive understanding of reconstructive decision-making.
Summary updated