Long-term survival in a fraction of patients with metastatic breast cancer who received consolidation therapy with high-dose chemotherapy and autologous stem cell transplant between 2000 and 2015: an EBMT registry-based study

Abstract

TO THE EDITOR: Breast cancer (BC) is the most common cancer and the leading cause of cancer mortality in women [1]. Despite advances in screening and treatment, nearly 12% of patients develop metastatic (M) disease. In the 80s and 90s, results of the early non-randomized clinical trials with high-dose chemotherapy (HDC) and autologous stem cell transplantation (ASCT) [2] demonstrated unusual improvements in patients with high-risk and MBC, creating great expectations with thousands of patients being treated with this modality, unfortunately in most cases given outside controlled trials [3]. In 2011, Berry et al. published the results of a meta-analysis, involving 866 women with MBC that evaluated HDC versus conventional chemotherapy (CC) [4]. The meta-analysis showed a statistically significant advantage for HDC in terms of PFS. However, this advantage did not translate into a survival benefit. Subsets of patients who may benefit from HDC were not identified due to the lack of information on biological parameters and the heterogeneity of HDC and CC regimens used. The meta-analysis did not include two prospective studies showing a potential survival benefit of HDC [5, 6]. This retrospective study analyzed the EBMT registry data regarding adult patients with MBC who underwent HDC and ASCT between 2000 and 2015. In the absence of robust prospective trials, the EBMT registry remains an important source to survey indications, outcomes, and clinical risk factors. OS and PFS were defined as time to death from any cause from the date of the first ASCT, and as the time from first transplantation to progression or death, respectively. OS and PFS were calculated through Kaplan–Meier (KM) analysis. A comparative analysis was conducted to select covariates associated with PFS by log-rank test. In Cox models, data were expressed as hazard ratio (HR), 95% CI, and p values. A total of 385 patients were identified, and 219 were eligible for OS and PFS analysis. Most of the patients were treated outside of clinical trials, and 184 patients have HER2 status reported. All transplants were performed using peripheral blood stem cells. Mortality within 100 days from transplant occurred in 12 patients (3.1%) due to transplant-related complications (TRM) or rapid disease progression; one secondary malignancy (ovarian cancer) was recorded; no haematological malignancies have been reported. The median age of patients at transplant was 47 years (range 40.9–53.6). The HDC employed included alkylating agents, mainly cyclophosphamide and thiotepa. Mitoxantrone, in combination with an alkylating agent, was utilized in 30.3% of patients. In 219 evaluable patients, the median OS was of 50 months (95% CI 43–70 months) and PFS of 18 months (95% CI 15–21 months). The OS and PFS at 5 and 10 years were 46%/29% and 14.2%/ 12.6%, respectively (Fig. 1a). PFS was unaffected by the type of HDC used, age at transplant, and HER2 status, whereas it was related to hormone receptor status. PFS was significantly worse in patients with a triple-negative pattern (excess risk: +72%, p= 0.01), in women who underwent HDC after 6 months from the diagnosis of metastasis (extra risk: +53%, p= 0.005) and in those with visceral or CNS metastases (excess risk: +48%; p= 0.01). Patients receiving HDC in stable disease or progressive disease had a significantly worse PFS than those receiving HDCT after obtaining a response to CC. Among patients transplanted in CR or PR, 15.4% and 14.4% were relapse-free at 5 and 10 years, respectively. HDC with ASCT produced a CR in 40 out of 190 patients (21%). The univariate analysis of PFS by status before and after HDC and by risk categories is given in Fig. 1b. However, the difference between CR–CR and new CR groups disappeared (HR: 0.73, 95% CI 0.39–1.40, p= 0.35) after data adjustment for potential confounders. Due to the associations between triple-negative pattern (yes/ no) and time from metastasis to transplant (<6/≥6 months) with PFS and because of the interaction between these two variables with age (≤50/>50 years), three risk groups (low, intermediate, and high-risk categories) were identified based on their median times to relapse. A low-risk included young women treated before 6 months from metastatic disease, irrespective of any triplenegative pattern (median survival time >700 days); intermediate risk, young women treated after 6 months or more, without triplenegative pattern (median survival time ranging from 400 to 700 days); high-risk, older women with a triple-negative pattern treated at various times, and younger women treated after 6 months with triple-negative pattern (median survival time <400 days). The KM analysis revealed significant differences in time to relapse between the three risk groups (Log-Rank test p < 0.001). Such a risk stratification was unaffected by data adjustment for potential confounders (Fig. 1c). In such a Cox model metastatic site and response patterns also maintained an independent relationship with PFS. While HDC continues to be part of the standard of care for a small number of mostly hematological cancers [7], it is no longer a treatment option for BC due to conflicting study results and