The association between age and lymph node status among patients with breast cancer after neoadjuvant chemotherapy: Eastern European population-based retrospective cohort study

Breast cancer (BC) remains a primary global health concern, with an increasing rate of individuals being elderly, i.e., diagnosed over the age of 65 (Howlader 2021). It is estimated that the incidence in elderly patients will reach 35% of all BC diagnoses by 2035 (Rosenberg et al. 2015). Within this population, both breast-conserving treatment (BCT) and mastectomy have been proven effective in terms of overall survival (OS) (Williams et al. 2022). However, the debate over the necessity of axillary procedures in elderly BC patients is still ongoing (Biganzoli et al. 2012). Both sentinel node biopsy (SNB) and axillary lymph node dissection (ALND) can result in complications and impaired quality of life. Furthermore, the long-term oncological benefit of these procedures remains unclear (Hughes et al. 2013, Vrancken Peeters et al. 2024). In 2016, the Society of Surgical Oncology guidelines advised against the routine application of SNB for women aged 70 and above who exhibit clinically negative lymph nodes (LN) and have early-stage hormone receptor-positive, HER2-negative invasive BC (http://www.surgonc.org/wp-content/uploads/2020/11/SSO-5things-List_2020-Updates-11-2020.pdf2020). However, there is limited data on oncological outcomes in elderly BC patients receiving neoadjuvant chemotherapy (NAC).

Although elderly patients are much more likely to be diagnosed with hormone-sensitive BC, they often present with contraindications to systemic chemotherapy due to comorbidities and frailty (Howlader et al. 2014, Plichta et al. 2020, Repetto 2003). Additionally, response rates to NAC in elderly patients, particularly in the Eastern European population, remain to be elucidated. A recent pooled analysis of eight randomized controlled trials (RCT) showed lower rates of pathologic complete response (pCR) in elderly BC patients compared with non-elderly individuals (12% vs. 21%) (Waldenfels et al. 2018) In this context, the ADVANCE pilot study focused on evaluating and tailoring NAC and adjuvant chemotherapy (AC) receipt in elderly BC patients (Freedman et al. 2023). Although the goal of the study was to provide preliminary results as a precursor to a large clinical trial, the endpoints were not achieved and the optimal systemic treatment for geriatric patients was not established.

The aim of the current study was to evaluate the impact of age on LN status, clinical outcomes, and OS in elderly BC patients after NAC. In addition, the influence of demographic and clinical factors on the incidence of LN metastasis was assessed.

The treatment of elderly cancer patients remains a complex and debatable issue. Given the absence of uniform protocols, surgeons and clinicians should focus on delivering appropriate and tailored care to the geriatric population in the era of multimodal treatment. Considering shorter life expectancy among elderly individuals, a therapeutic approach should include maintaining acceptable quality of life and minimizing functional impairments. The current study was important, as it sheds new light on adjusting surgical treatment among elderly BC patients undergoing NAC. Although the two study groups showed comparable response rates to systemic treatment, different clinical conclusions could be drawn. In the non-elderly group, a pCR was associated with decreased odds of LN involvement, while no such association could be shown within the elderly group. Furthermore, LN status had no influence on the OS within the elderly group, with higher cT stage being the sole factor of decreased risk of death. Nonetheless, this counterintuitive observation should be interpreted with caution, as it may reflect selection bias. Elderly patients with larger tumors who were selected for NAC and surgery possibly represented a more fit subgroup, while some individuals with smaller tumors yet worse overall health may have had unfavourable outcomes due to non-cancer reasons.

As shown in the study from Memorial Sloan Kettering Cancer Center, even though elderly BC patients are under-represented in RCTs, their clinical outcomes after NAC receipt are similar to younger patients (Williams et al. 2022). The conversion rate from locally advanced setting to BCT-eligible was comparable between the age groups (72% vs. 74% in elderly and non-elderly individuals, respectively). Furthermore, women aged ≥ 70 who converted to BCT-eligible post-NAC were more likely to undergo BCT than younger patients (Williams et al. 2022). Consistent with these findings, in the current study, there were no differences between the response to NAC between elderly and non-elderly patients. In addition, elderly patients had a higher proportion of < 90% tumor regression compared with younger patients (52.7% vs. 45.4%). Of note, the baseline cT distribution did not differ significantly between elderly and non-elderly patients, which indicates that the differences in ypT were more likely the result of reduced chemosensitivity in older patients, leading to lower pCR rates. However, this interpretation should be viewed cautiously, as our dataset does not include information on chemotherapy-related toxicity, dose reductions, or treatment intensity, which may also contribute to lower pathological response rates in older patients.

LN status remained unassociated with pCR in elderly patients. Interestingly, despite lower pathological response rates, OS did not differ between elderly and non-elderly patients, suggesting that NAC response does not directly translate into long-term survival in this population. This may suggest that further prospective research on the benefits of NAC administration in elderly patients in the context of axillary LN interventions is needed to draw firm conclusions.

In German pooled analysis of individual patient data from eight prospective RCTs, geriatric patients had significantly larger tumors (stage T4a-d) and LN involvement (LN + 1–9) compared with younger patients (Waldenfels et al. 2018). Similarly, in the current study, elderly patients more often had higher ypT stages and nodal metastases after NAC compared with non-elderly patients. Consequently, BCT and SNB were used less frequently among the geriatric cohort. Of note, pathologic staging after NAC in elderly patients did not affect survival outcomes, in contrast to non-elderly patients, in whom LN involvement was associated with an increased risk of death.

A “de-escalation trend” suggesting omitting ALND in BC patients has recently gained popularity. As shown in ACOSOG Z0011 and AMAROS trials, omission of ALND, even in individuals with diagnosed LN metastases, has no negative consequences regarding OS (Ortega Exposito et al. 2021, Poodt et al. 2018). However, the vast majority of studies involving elderly BC patients include patients undergoing upfront surgery and adjuvant radiotherapy, without NAC receipt. Nonetheless, it seems that the conclusions of previous research are consistent with the results of the current study, which suggests that within elderly BC patients, LN status does not affect survival rate, despite generally higher clinical staging. This finding further supports ongoing efforts toward axillary de-escalation in well-selected elderly patients, although prospective validation is required.

Notably, the higher proportion of luminal tumors receiving NAC in our cohort reflects real-world practice in an Eastern European setting rather than a deviation from contemporary guidelines. Similar to our centers, Barbieri et al. demonstrated that HR+/HER2 - tumors - particularly luminal B or clinically node-positive cases - frequently receive NAC to facilitate breast-conserving surgery or to achieve axillary downstaging, despite overall lower pCR rates (Barbieri et al. 2021). Likewise, Torrisi et al. emphasized that NAC remains an appropriate strategy in selected luminal tumors with high proliferation indices or unfavorable tumor-to-breast ratios, where downsizing may meaningfully alter the surgical plan (Torrisi et al. 2021). Because genomic assays were not widely available during the early years of our study, multidisciplinary teams often relied on these same clinical features when recommending NAC. This context explains the higher NAC utilization among luminal patients observed in our cohort.

The association between obesity and adverse pathological features in breast cancer provides important context for our findings. Excess adiposity has been shown to reduce the likelihood of achieving pathological complete response after neoadjuvant chemotherapy, as demonstrated in a recent meta-analysis by Wang et al. (Wang et al. 2021) In parallel, a dose-response meta-analysis including over 52,000 patients found that increasing BMI is independently associated with a higher risk of lymph node metastasis at diagnosis (Wang et al. 2020). These established relationships support the relevance of obesity within our analysis and may help explain the observed link between higher BMI and residual nodal disease following NAC in reported cohort.

The findings of the current study should be addressed with certain limitations. Firstly, aggregation of all breast cancer subtypes into a single analytical cohort instead of subtype-specific evaluation. Although molecular subtype was included as a covariate in the regression models, residual confounding arising from the marked biological and therapeutic heterogeneity across subtypes cannot be fully excluded. Robust subgroup analyses were not feasible given the retrospective design and the limited sample sizes within individual molecular subtypes, which precluded adequately powered stratified analyses. As a retrospective study conducted in two centers, our results are also subject to selection bias, especially regarding which elderly patients were offered NAC and surgery. Secondly, the cohort was not analyzed based on socioeconomic status and comorbidities that could influence the results, particularly within the elderly group. Another important limitation is the relatively small number of elderly patients in the ypN + and ypN– subgroups. This reduces the statistical strength of comparisons and may have contributed to the lack of significant associations with survival. Furthermore, The high proportion of elderly and luminal patients receiving NAC likely reflects treatment selection bias inherent to retrospective datasets. Multidisciplinary Team decisions were influenced by factors not included in the dataset (patient frailty, desire for breast conservation, imaging characteristics) or locally evolving practice patterns - limiting direct comparability to current guideline-based cohorts. Notably, 6.4% of individuals underwend ALND node dissection without prior SNB. As a result, some patients may have had undetected nodal disease at baseline, potentially leading to an overestimation of the true proportion of cN0 cases. Cause of death was not available in the current study, preventing differentiating BC–specific mortality from deaths due to other causes. This limitation is especially relevant in elderly population, in whom competing comorbidities may substantially influence OS. Finally, the elderly and non-elderly groups were numerically unbalanced, which reflects the age distribution of patients receiving NAC in real-world practice; this imbalance may limit the comparative power of subgroup analyses and should be interpreted accordingly.

Although presented outcomes cannot be directly translated into clinical practice, it is crucial to note a relatively large sample size of elderly BC patients receiving preoperative systemic treatment followed by curative-intent surgery. These limitations should be taken into account when interpreting our results, and confirmatory prospective studies are needed to validate our observations.

In conclusion, our study underscores the possible benefits of NAC within BC patients based on age. While LN status did not impact long-term prognosis among elderly individuals, nodal involvement after NAC was a poor prognostic factor among non-elderly patients. This seemingly paradoxical finding in the elderly cohort may be explained by selection bias. Patients with higher cT tumors who were offered NAC and surgery likely represented a fitter subgroup, whereas some elderly patients with smaller tumors but poorer overall health may have experienced worse survival due to non-cancer causes.