Breast conserving treatment is the standard of care for early stage breast cancer (26). RT is a mainstay of this conserving approach, not only allowing a threefold reduction in LR but also improving OS (3,27). Patients aged 70 years and over, who are in good health condition, have a median life expectancy about 15 years and half of them will live much longer.
Elderly breast cancer represents a public health challenge, which will become more and more critical with the upcoming increasing life expectancy during the next decades and the impact of a potential acceptance of breast cancer screening for patients older than 75. For elderly patients presenting with early breast cancer, choosing the appropriate adjuvant treatment remains difficult, and we need to promote more accurate strategies taking into account oncological outcome and treatment toxicity as well as quality of life and health care expenditure.
In order to find an acceptable compromise between no breast irradiation, as suggested in randomized trials (7–12), and 3–5 weeks of WBI, the concept of APBI emerged and has been presented as a good option specifically for the elderly population with breast cancer at low-risk of LR (17,18). The rationale for investigating partial-breast radiotherapy is based on the fact that after breast-conserving surgery, most LR occur at or near the primary site of the cancer (tumor bed) (28,29). Indeed, irradiation is limited to the higher risk area of LR. By increasing the radiation fraction size and decreasing the target volume and consequently decreasing the volume of irradiated normal tissue, APBI represents an alternative for the local treatment of breast cancer, and our extended follow-up enables us to look at the long-term impact of this approach.
Many techniques of APBI have been developed: HIBT, external breast RT and intraoperative radiation therapy. HIBT is one of the first APBI technique used and has the longest follow-up (30,31). The two last trials which reported the comparison between external partial breast RT to WBI confirmed the non-inferiority of the APBI technique in the RAPID trial, (but with a significant deleterious impact in terms of cosmetic outcome) (32,33) but not in the NSABP B-39/RTOG 0413 trial (34). Two trials of intraoperative radiotherapy reported higher rates of LR compared with conventional WBI (35,36). The GEC-ESTRO APBI phase III compared WBI versus APBI for selected primary early-stage breast cancer (low and intermediate-risk groups) (37) and the 5-year LRFS was 98.6%, which is similar to our 5-year LRFS of 98.1% previously described (19). With a 12-year median follow-up, Polgar et al. (38) reported a 5-year LR rate of 4.7% with 77% of good/excellent cosmetic results.
This update confirms the outcome of our cohort while HIBT remains as an attractive technical option (19). In our study, regarding the oncological outcomes of the whole cohort and taking into account that in specific circumstances, we were led to propose APBI even for patients belonging to the APBI GEC-ESTRO high-risk group, the 8-year LRFS, MFS, SS and OS rates were 96.7%, 95.7%, 96.7% and 71.8%, respectively. Focusing on local control, Polgar et al. reported a 10-year LR rate of 9.3% but in a lower risk breast cancer population and younger cohort (median age: 56 years) (38). Our results lead to consider that, in an elderly population presenting with an early breast cancer, APBI could be reasonably proposed for intermediate and even high-risk patients in order to achieve optimal local control rate. Comparing with external beam APBI phase III randomized trials, we observed similar results in terms of local control. With a median follow-up of 10.2 years, the NSABP B-39/RTOG 0413 (34) reported a LRFS rate of 95.4%. With a similar 8-year median follow-up, the LRFS rate reported in the RAPID trial was 97% (32).
In our cohort, we observed that the majority of the patients died from other causes than cancer (Fig. 2). The difference between 8-year SS (96.7%) and OS (71.8%), can mainly be explained by our very old population (median age: 81.7 years) in which a lot of competing mortality factors come into play.
Regarding oncological prognostic factors, according to the GEC-ESTRO APBI classification, we reported 8-year LRFS rates of 98%, 87.5% and 100% for low-, intermediate- and high-risk groups respectively. In univariate analysis GEC-ESTRO APBI classification was not significant for LRFS (p = 0.406), MFS (p = 0.616), SS (p = 0.436) and OS (p = 0.408). Although based on a small number of events, it seems that for elderly patients this classification may not be as accurate as for the younger ones, and APBI could be discussed as a treatment option for selected high-risk patients. In our study, even if the molecular classification (39) was considered as a significant prognostic factor for OS (p < 0.0001), SS (p = 0.007) and MFS (p = 0.009) but not for LRFS (p = 0.586), those results must be considered with caution because of the very low number of triple negative patients associated with a related high rate of oncological events (40,41). Nevertheless, molecular classification could impact on local control and has not been considered directly in the GEC-ESTRO APBI classification. Although this classification takes into account HR status, Her2 status and Ki67 are not considered. Considering molecular classification combined with GEC-ESTRO APBI classification could likely lead to a more accurate patient selection for APBI.
In our study, the maximum complication rate observed after APBI, was 88.4% (61 patients), 11.6% (8 patients), grades 1 and 2 respectively. No grade ≥ 3 was observed. Other trials evaluating different partial breast irradiation methods have reported good toxicity profile but more grade ≥ 2. The RAPID trial (32) reported 32% of grade ≥ 2 and 4.5% of grade 3 while the GEC-ESTRO trial (42) reported 23.3% of cumulative incidence of grade 2 or worse late toxicity. In the NSABP B-39/RTOG 0413 (34) adverse events were worst with a highest toxicity grade reported from APBI of grade 1 in 40%, grade 2 in 44%, and grade 3 in 10%. Regarding cosmetic outcome, the rates of excellent and good cosmetic outcomes were 57.8% and 38.6% respectively, while a fair result was observed for 2 patients (1.8%).
In phase III randomized trials comparing surgery plus endocrine therapy with or without adjuvant WBI, a significant over-risk of LR without breast irradiation was reported (9,11). Due to the deleterious impact of aromatase inhibitors on the quality of life (43), it is currently discussed to promote adjuvant breast irradiation without endocrine therapy in the elderly with low-risk breast cancer (44,45). In our population study, 5 patients refused and 12 patients stopped the treatment prematurely due to poor tolerance. For selected patients, APBI could also be considered as a viable alternative to the omission of adjuvant RT which has been considered for elderly patients, as it drastically reduces the number of transportations, alleviates the treatment related constraints, in particular for elderly patients with frequent comorbidities, without compromising local control. Currently, some protocols wants to get further proposing single-fraction HIBT with promising results (46,47).
In a time of cost saving for patients and health care systems, Shah et al. (48) showed in a cost-efficacy study that compared with WBI using three dimensional conformal RT and intensity-modulated RT, every APBI techniques allow for 1000 treated patients, cost savings between $0.7 and $6 million and between $5 and $14.9 million, respectively. Lanni et al. (49) confirmed that APBI using brachytherapy techniques was less costly than conventional WBI with a standard boost.
The limitations of the present study are mainly represented by its retrospective status, the small number of patients and (fortunately) the very few numbers of oncological events that did not allow performing an accurate statistical analysis for prognostic factors.