Radiation oncology: Do we need a boost for every patient?

After breast preservation, whole-breast radiotherapy is usually indicated. The addition of a supplementary irradiation dose to the tumor bed, known as boost, can be individualized in most patients. In this article, we discuss how and why.

Breast cancer (BC) radiotherapy (RT) is an essential part of BC treatment, improving local control and overall survival.¹ Whole-breast RT is used almost systematically after breast-conserving surgery (BCS)¹ and an additional dose to the tumor bed, called boost, is often used to improve local control.² However, RT also comes at a price of acute and, importantly, late toxicity,³ which has to be weighted over oncological benefits. The challenge of modern BC RT today is to be tailored to the risks and needs of the individual patient, in a spectrum between intensification for high-risk patients and de-escalation for those with low risk.⁴

For these low-risk patients, who represent the majority of cancer diagnoses since the introduction of screening,⁵ the need to boost the tumor bed with an additional dose can be challenged. De-escalation has indeed been advocated since the ’70s and ’80s but has never been more personalized than in recent years.⁶ The rationale of boosting the tumor bed has been based on the observation that within 0.5 to 4 centimetres from the initial tumor, BC cells can be found within mastectomy specimens.^7,8

As local recurrence (LR) patterns of many studies have shown that 44% to 80% of LR occur within the site of the initial surgery⁹ and the optimal dose to deliver to the tumor bed was not known, EORTC designed a large phase III study, recruiting patients with T1–2 N0–1 tumors treated with BCS and a resection margin of 1cm between the years 1989 and 1996.

The study’s first report showed a benefit in local control with the addition of boost (HR for LR: 0.59 for all patients, 0.66 for those with no adjuvant systemic treatment and 0.4 for those with adjuvant systemic treatment).¹⁰ Of note, in that study, systemic treatment was dramatically suboptimal as compared to current standards.¹⁰ Already in this report, the effect of age on the magnitude of benefit from boost was clear, while outcomes at 20-year follow-up confirmed the boost benefit in local control and the absence of benefit in overall survival (OS).² Benefits in local control decreased with age, remaining statistically significant, while risk of fibrosis, although overall low, increased with age, for patients older than 41 years.²

A central pathologic review of 1/3 of cases in this study revealed a benefit of the addition of boost depending on age and grade, while no benefit was observed when resection margins were positive. The benefit persisted when margins were negative.¹¹ A central review undertaken at 18-year follow-up revealed a cumulative incidence of LR of 34%, 14% and 11% for patients aged 40, 41 to 50 and >50 years, respectively.¹² Presence of ductal carcinoma in situ (DCIS) or not was associated with LR-cumulative incidence rates of 18% vs. 9%, respectively. The use of a boost decreased 20-year LR incidence from 31% to 15% for high-risk patients (younger than 50 years with DCIS).¹² At multivariate analysis, age and DCIS remained statistically significant, with presence of DCIS remaining important over time, while grade exhibited its prognostic importance mainly in the first two years.¹²

This evidence challenges somehow existing guidelines¹³ and common practice, as discussed by Reht et al.,¹⁴ in an insightful commentary published in 2018. For example, they argue that omitting a tumor bed boost dose in patients >70 years with low-risk disease (hormone-receptor positive and low grade), is a recommendation derived from studies of omission of RT in the elderly, where boost, however, was only delivered in 16% of patients.¹⁵

Fig. 1: Locoregional control, distant control and overall survival (OS) in the BIG 3–07/TROG 07.01 trial. Adapted from Chua BH et al.¹⁶

Another debated issue was the use of boost for DCIS. Thanks to the publication of the BIG 3–07/TROG 07.01 study,¹⁶ we know now that a boost will improve local control at five years from 93% to 97% for all subgroups of patients with non-low risk DCIS (Fig. 1), but will slightly increase grade 2 and 3 toxicity and impair quality of life (QoL).¹⁷ It should thus be the object of a shared decision-making procedure.¹⁶

Another debatable point has been a hypofractionated boost and/or the use of a boost after a hypofractionated regimen for whole-breast irradiation. This is due to the fact that pivotal hypofractionation studies did not include a boost,¹⁸ although the TROG 07.01 study also addressed the question of hypofractionated whole-breast RT for DCIS and showed non-inferiority.¹⁶ It is important to remember that in the subgroup of patients who received a boost in the combined analysis of the START trials¹⁹ the analysis did not show any impact of boost in favour of the normofractionated or hypofractionated regimens.¹⁹ Studies such as IMPORT-low²⁰ and IMPORT-high²¹ and, importantly, the NRG/RTOG1005 trial²² have confirmed that a simultaneously integrated boost can be delivered, reducing the number of fractions and improving comfort.

Dose matters, as far as toxicity is concerned. As shown in the young boost trial within patients up to 50 years old, a boost of 26Gy improved locoregional control at ten years from 4.4% to 2.8%, but also moderate fibrosis from 27% to 48%.²³ Importantly, one must not forget that the risk of breast induration, slightly increased by the use of boost, mainly depends on the volume of the irradiated breast overall at higher doses, especially in patients aged 65 years and over.²⁴ This risk is also doubled by smoking, irrespective of volume and age.²⁴

Trends over time show indeed a more personalized use of boost within all patients, as well as, specifically, those with low risk and early-stage disease.²⁵ Within this context and given the granularity of information available today on the use of a boost, it becomes clear that many decisions on its use must be taken after a shared-decision making consultation, where patients’ needs, priorities and concerns are heard and integrated in the therapeutic process.²⁶ This remains probably the best way to improve care and bring clinical research’s advents to our patients.