Breast-conserving surgery followed by a whole-breast postoperative RT (WBRT) is widely considered to be the current standard of care for patients with early breast cancer. This treatment leads to an excellent local tumor control with local recurrence rates around 6% after 10 years, particularly in the very-low risk subgroup 1 . This classical 5–7 weeks WBRT is simple and safe in the large majority of patients but may bring some local early and late side effects. In addition, it is frequently inconvenient for women, namely in the elderly, who sometimes prefer to omit it assuming the risk of local recurrence. During the last decades, through numerous observational series and large clinical trials, more than 80% of the local relapses occurred in the same quadrant in this clinical situation 2 .

On this basis, reducing both the treatment volume and duration was considered to be an innovative approach and was developed under the name of accelerated partial breast irradiation (APBI). Intraoperative radiotherapy (IORT) is one of them that offers the advantages of an excellent delineation of the tumor bed under visual control, a very good dose homogeneity, and a high normal tissue sparing 3 4 .

Alongside our IORT experience delivered as a boost 5 , we started in 2004 the RADELEC phase II trial evaluating IORT as a sole treatment in very low-risk patients. We report here the long-term results of safety, cosmesis, and carcinologic events.

This study was conducted in accordance with the declaration of Helsinki and the local institutional review board. All patients provided written informed consent.

The updated publication of the collaborative meta-analyses based on individual patient data confirmed that WBRT halves similarly local recurrence rates in the different subgroups of patients and reduces the breast cancer death rate by about a sixth 1 .

The need of WBRT was however debated since the publication of the CALGB 9343 trial estimating a low risk of recurrence after conservative surgery in patients older than 70 years presenting HR+ small tumors and only adjuvant tamoxifen as adjuvant treatment 11 . These results were updated at the annual meeting of the American society of clinical oncology (ASCO) in 2010 and showed an increase risk of local recurrence when adjuvant radiotherapy was avoided. At a median follow-up of 10.5 years, 98% of the radiation group and 92% of the tamoxifen-only group were recurrence-free confirming the local risk of avoiding adjuvant irradiation. In addition, the results of the NSABP-21 12 confirmed that the absolute risk of in-breast recurrences of primary small tumors less than 1 cm is not low enough to spare patients the need for WBRT.

Nevertheless, the standard of 5 to 6 weeks WBRT is no longer the optimized strategy as it appears long and binding in this clinical situation. Indeed, the concept of accelerated whole-breast irradiation recently showed a risk of local recurrences at 10 years of 6.7% among the 612 women assigned to standard irradiation as compared with 6.2% among the 622 women treated in 3 weeks 13 . At the San Antonio Breast Cancer Symposium 2012, the START B trial 14 was updated by Yarnold et al. and confirmed 5.5% and 4.3% of local recurrences at 10 years in the standard and accelerated WBRT, respectively.

Considering that 80% of the local breast recurrences occur within the tumor bed 15 , other strategies attempted to reduce the irradiated volume when accelerating the overall treatment time 16 . IORT procedure represents one of the possibilities offering a one-fraction treatment in a limited volume during the primary surgery. Our results confirmed with a long-term follow-up that exclusive partial breast using IORT is feasible for early-breast cancer with an absolute risk of carcinologic events extremely low in much selected patients, namely in the elderly. In that condition, using IORT for partial irradiation with a 21-Gy fraction has the major and unique advantage of a “one-shot” procedure including surgery and radiotherapy at the same time. Extending the duration of surgery, from 20 to 50 minutes, permits to avoid 5 to 6 tiring weeks of external radiotherapy, one of the reasons that encourage patients to decline adjuvant radiotherapy or ask for a mastectomy 17 . This argument is reinforced by the proportion of the patients receiving the recommended adjuvant radiotherapy, less than 75% after seventy and even less than 50% after eighty 18 . IORT may therefore represent an alternative considering the shortness and simplicity of the technique providing that this treatment is done in expert hands.

Our updated results confirm with a long-term follow-up that IORT, as an accelerated partial breast irradiation technique, is suitable for selected patients and could be a better compromise regarding tamoxifen alone in adjuvant setting. However, the local recurrence rates at 6 years, for this very favorable group of patients, seems quite high (9%), especially given the fact that this subgroup of patients with indolent disease will continue to recur in the next 10 and 15 years, for whom rates of 6% at 10 years have been approximated. Therefore, longer follow-up is needed to draw meaningful conclusions.

Alongside our experience in IORT delivered as a boost 4 5 , we decided many years ago to extend this concept to a specific very low-risk population (i.e. age ≥ 65 years old, tumor size < 2 cm, non-lobular carcinoma and estrogen receptor positivity) for a unique and exclusive treatment. We excluded BRCA1 or 2 carriers and extensive in situ carcinoma. Since then, the American and European societies for therapeutic radiation oncology (ASTRO and ESTRO, respectively) provided a consensus statement for the use of accelerated partial breast irradiation based on current published evidence and completed by expert opinion 19 20 . The patient selection in the present study is perfectly concordant to these consensuses and reinforces the necessity to respect all the stringent criteria for further studies 21 .

The choice of electrons for this partial breast radiotherapy was based on our local experience in this technique 4 5 . We showed that electrons allow an homogenous dose, spare the skin and give time to surgeons for a post-resection reconstruction of the cavity leading to very good cosmetic results. Orecchia et al. presented the 5-year local recurrence rate of patients included in the ELIOT trial (electrons’ technique) at the last World Congress of Brachytherapy in May 2012. The authors observed 3.6% of local recurrence but longer follow-up is warranted to draw definitive conclusions.

Recently, the TARGIT-A trial 22 was published comparing standard WBRT to a single 20-Gy fraction delivered intraoperatively but with a 50-kV system. The 4-year local relapse rates were not inferior in the IORT arm (0.95 and 1.20% in the WBRT and IORT arms, respectively). Mature results were presented at the last San Antonio Breast Cancer Symposium 2012. The investigators continued inclusions after the first publication (from 2232 to 3451 patients) that renders difficult long-term results. The 5-year risks for local recurrence in the conserved breast for TARGIT vs WBRT were 3.3% (95% CI 2.1-5.1) vs 1.3% (95% CI 0.7-2.5). Nevertheless, TARGIT had similar results to WBRT 2.1% (1.1-4.2) vs 1.1% (0.5-2.5) in the pre-pathology subgroup (concomitant surgery and TARGIT) that reinforces the idea that delayed procedure by reopening the wound has to be abandoned. Final data are still pending for publication.

As local recurrences may occur after a long delay, final assessment of kV-IORT will be definitely valid after sufficient follow-up from large international prospective randomized trials.

Frozen section is, for sure, one limiting aspect of any intraoperative procedure, as the definitive pathology report may contradict the biopsy. This technique requires therefore a very close involvement of the pathologist, the surgeon, and the radiation oncologist.

Even if the cosmetic results were evaluated as good, structural changes in the tumor bed after IORT may require a learning curve for the radiologist in order to avoid iterative biopsies 23 . The patient assessments seem extremely important before considering this technique as a standard in daily practice. The use of 50-kV IORT will reduce the risk of late fibrosis and cytosteatonecrosis as the need of dissection is highly less compared to IORT with electrons (71% in the current trial). Indeed, the sphere of Intrabeam fill in the surgical area whereas the tissue surrounding the excision cavity is mo1bilized and approximated by sutures to bring it into the RT planning target volume with electrons.

In contrast, IORT has several advantages and some teams 24 emphasize that using IORT for adjuvant breast radiotherapy may reduce the estimate of second-cancer risk. Compared to classical external WBRT or accelerated partial breast external irradiation, this technique delivers the lowest dose to the controlateral breast, homo-, and controlateral lungs and spine.

Finally, IORT reduces the cost of adjuvant breast radiotherapy and a medico-economic prospective study is still ongoing in France to evaluate the impact of this treatment on healthcare resources and public health.

In conclusion, our results confirm that IORT given as a sole treatment during breast-conserving surgery is a reliable alternative to conventional postoperative fractionated radiation. This one-stop treatment reduces patient effort and limits the use of health care resources. It could be considered as a standard in a selected population with very-low risk of local recurrence but performed by multidisciplinary expert hands. Patients’ assessments strongly improve long term evaluation of this technique.

WBRT: Whole-breast postoperative RT; APBI: Accelerated partial breast irradiation; IORT: Intraoperative radiotherapy; IBRT: Ipsilateral breast tumor relapses; EIC: Extensive intraductal; ER: Estrogen receptor; PR: Progesterone receptor; QI: Quality index; DFS: Disease-free survival; OS: Overall survival.

All authors declare that they have no competing interests.

CL, JBD, and MG conceived the study, collected data, and drafted the manuscript. AP, PF, and NA collected data. DA participated in coordination and helped to draft the manuscript. CL, PR, MG, and JBD participated in the design of the study and assisted in data collection.SGB performed the statistical analyses. DA provided mentorship and edited the manuscript. All authors have read and approved the final manuscript.