Breast invasive cancer adjuvant EBRT hypofractionation whole breast

Target verification notes

• Consideration should be given to dose delivered from verification imaging and whether this needs to be accounted for in treatment planning, particularly with respect to critical OARs.

Efficacy

The evidence supporting this protocol is provided by three large phase III multi-centre international randomised trials with 10 year follow-up results, and the Cancer Australia guidelines systematic review update.^r

Canadian trial

A Canadian study^r by Whelan et al randomized 1234 women with completely excised invasive breast cancer (pT1-2, N0, M0) to 50Gy/25# or 42.4Gy/16# treating 5 fractions per week. The 10-year cumulative incidence of local recurrence was 6.7% among women in the conventional group versus 6.2% among women assigned to the hypofractionated group (absolute difference 0.5 percentage points, 95% CI -2.5 to 3.5). The 10 year probability of survival was similar between the two groups (84.4% conventional vs. 84.6% hypofractionated group). There was also no significant difference in the rate of good or excellent cosmetic outcome between the conventional and hypofractionated groups.

UK START-A and START-B trials

The 10-year update^r on UK START-A^r and START-B^r trials also confirmed the efficacy and safety of hypofractionated radiotherapy in women with completely excised invasive breast cancer (pT1-3a, N0-1, M0). START-A randomized 2236 women to receive 50Gy/25# compared with 41.6Gy/13# or 39Gy/13# all over 5 weeks. START-B randomized 2215 women to 50Gy/25# or 40Gy/15# with both regimens involving 5 fractions per week.

In both trials, there was no significant difference in 10 year locoregional control between the conventional and hypofractionated regimens. Both trials also showed a significant reduction in late normal tissue effects in the hypofractionated groups, with the START-A trial showing less breast induration, telangiectasia and oedema in the 39Gy group, and START-B showing less breast shrinkage, telangiectasia and oedema in the 40Gy group than the 50Gy group.

For 10 year all-cause mortality, in the START-A trial there was no significant difference between 41.6Gy and 50Gy (18.4% vs. 19.8%, p=0.74) or the 39GY and 50Gy groups (20.3% vs. 19.8%, p=0.69) In the START-B trial, the hypofractionated group had a significant reduction in the rate of distant metastases, which contributed to a small reduction in the rate of all-cause mortality in this group compared with conventional fractionation (15.9% vs. 19.2%, p=0.042). However, this reduction cannot be readily explained in terms of radiotherapy-related factors. There were significantly fewer 10 year distant relapses in the 40Gy group compared with the 50Gy group (12.3% vs 16%, p=0.014), which contributed to the significantly higher rate of OS in the 40Gy compared with the 50Gy group.

Sub-groups

Planned sub-group analysis in the Canadian trial showed that treatment effect was similar regardless of age, tumour size, oestrogen-receptor status, or use of cytotoxic chemotherapy. The Canadian trial did not utilise a tumour boost because study was initiated at a time prior to publication of studies supporting boost irradiation.^r The Canadian trial excluded patients with node positive disease and large breast (separation >25cm).

Post-hoc subgroup analyses of the combined hypofractionated regimens versus conventional fractionation in the START-A and START-B trials showed that the treatment effect was not significantly different irrespective of age, breast size, tumour grade, axillary node status, cytotoxic chemotherapy or tumour bed boost radiotherapy. However, only a small proportion of patients in the START-A and START-B trials received regional nodal irradiation, or contemporary adjuvant chemotherapy.

A number of studies showed no difference in rates of local relapse for grade 3 breast cancers receiving hypofractionation compared with conventional fractionation.^rrr Bane et al^r found that molecular subtype predicted for local recurrence in women enrolled in the hypofractionated whole breast irradiation trial after a median follow up of 12 years. However, tumour grade, molecular subtype and hypoxia did not predict response to hypofractionation suggesting that patients with node-negative breast tumours of all grades and molecular subtypes may be safely treated with hypofractionated regimens. This is supported by the recent Cancer Australia guidelines.^r

Toxicity

A summary of the most clinically significant toxicities associated with this protocol are included in the table below.

*Concerns have been raised about doses to the heart with hypofractionated schedules. The heart is sensitive to radiation whatever fractionation is used, with no dose threshold for adverse effects. START-A and B showed no increased cardiac events with hypofractionated regimens, however, follow-up in these studies is considered too short to assess late cardiac events. The START-A and B trials excluded patients with pre-existing heart disease from the analysis – other studies have demonstrated that the absolute effect of radiotherapy on heart is greater in patients with established cardiovascular system risk factors.^r The risk of cardiovascular disease in breast radiotherapy patients increases with longer follow-up beyond 10 years.^rr In one of these 2 retrospective reviews patients were treated with 4.3Gy per fraction.^r

Appelt et al^r examined the predictions of the linear quadratic model on the estimated fraction size-corrected dose to heart for four hypofractionated regimens used in START-A, START-B and Canadian trials and conventional fractionation for a range of a/ß ratios. The 40Gy/15#, 39Gy/13# and 42.5Gy/16# regimens were estimated to have lower dose to heart, adjusted for fraction size using the linear quadratic model, compared to conventional fractionation. However, 41.6Gy/13# was estimated to result in higher dose to heart if heart a/ß=1.