It is well accepted that the use of adjuvant radiation in the treatment of early breast cancer has been very successful in enabling a breast conserving approach for the majority of early breast cancer. Since the 1980s, a large number of randomised trials have confirmed the safety of this approach and have documented not only the loco-regional control benefit but also recently confirmed survival benefit and long-term safety of adjuvant radiation therapy for breast cancer. 

This prompted the National Cancer Institute in the US to issue a strong recommendation in 1990 in support of a breast-conserving approach for early breast cancer, stating: “Breast conservation treatment (BCT) is an appropriate method of primary therapy for the majority of women…and is preferable because it provides survival equivalent to total mastectomy and axillary dissection while preserving the breast.”

Numerous randomised studies have confirmed equivalent outcomes for BCT and mastectomy for early breast cancer. The Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) has reported on the results of 10,906 in 17 randomised trials of BCT with a minimum follow up of 9.5 years.¹ The addition of radiation treatment resulted in a 21.7% reduction in 10-year local recurrence (32% without radiation versus 10.3% with radiation), a 5.4% reduction in 15-year breast cancer mortality (35.9% versus 30.5%, respectively; p = 0.0002), and a 5.3% reduction in 15-year overall mortality (40.5% versus 35.2%, respectively; p = 0.005).¹

The recent confirmation that improvement in local control results in an overall improvement in survival is an important finding as, up to the late 1990s, the view was that breast cancer was largely a systemic disease and local therapy did not influence long-term outcome. This previously held view has now been shown to be incorrect and optimum local therapy does influence survival, not only in breast cancer but also in numerous other solid tumour sites. 

More recent trials of adjuvant radiation therapy have also confirmed that the rate of local relapse continues to fall and, internationally, the risk at 10 years of local relapse should be considerably less than 10%. 

Indeed, for recently treated patients the rate of local recurrence is now reaching 5% at 10 years (approximately 0.5% per year after treatment). This low level of expected recurrence is a good bench mark for modern breast centres and reflects not only better quality radiation therapy with improved dosimetry and appropriate use of boost (higher dose to tumour bed site), but also better quality surgery with margin assessment and pathological review.

Impact of a higher radiation dose on local control 

Bartelink et al, in one of the best quality modern radiation oncology trials reported in the literature, confirmed that increasing the dose to the tumour bed improves local control in all age groups but more significantly for young women (≤ 40 years of age).²

This study (the EORTC 22881-10882 trial) investigated the long-term impact of a boost radiation dose of 16Gy on local control, fibrosis, and overall survival for patients with stage I and II breast cancer who underwent breast-conserving therapy. A total of 5,318 patients with microscopically complete excision (with 1cm margin of excision) followed by whole-breast irradiation of 50Gy were randomly assigned to receive either a boost dose of 16Gy (2,661 patients) or no boost dose (2,657 patients), with a median follow-up of 10.8 years. 

Irradiation of the whole breast was performed using two tangential megavoltage photon beams (high-energy x-ray or tele-cobalt), with the boost dose of 16Gy delivered with electrons or tangential fields administered in eight fractions, or with an iridium-192 implant at a dose rate of 0.5Gy per hour. 

Strict quality assurances, as mandated in the trial design, included central pathological review, which was an important addition to the trial and reported on risk factors for local recurrence and influence of radiation therapy on outcome as well as long term complications and cosmetic results.

Bartelink et al reported that the boost of 16Gy in eight fractions reduced local recurrence from 10.2% to 6.2% (p < 0.0001, HR 0.59). The greatest reduction was seen in women aged 40 years or less (from 23.9% to 13.5%). 

As a result of this study, it is now standard of care to deliver a boost post standard conventional tangent radiation in all patients aged 60 years or younger, even with negative margins, and should be considered for older patients with close margins or other high risk pathological findings.

Rationale for hypofractionation

Given the confirmed benefits of standard fractionated radiation therapy for early breast cancer, is there a rationale for changing this safe and well-established radiation schedule? The addition of a 16Gy boost now means that the radiation course runs for 6.5 weeks, which can be difficult for many patients to access. Internationally, up to 30% of breast cancer patients do not choose breast-conserving therapy primarily because of access to radiation centres and length of time for treatment. Shortening the treatment course can have secondary benefits in reducing waiting times for busy radiation centres and can improve convenience for patients.

Before recommending a practice-changing approach for early breast cancer the following needs to be robustly confirmed:

• No detriment in outcome with respect to local control and survival rates
• No increase in complication rates
• No detriment in cosmetic results.

Significance of local regional recurrence

The European Organisation for Research and Treatment of Cancer (EORTC) pooled four studies from the years 1980-1999, which reported on the long-term results of local recurrence in breast cancer. Local regional recurrence was the strongest prognostic factor predicting lower overall survival (HR 5.01).  

The research showed that late local recurrence > 5 years also predicted with a HR 3.98 for poorer overall survival. With this finding, at least 10-year follow-up data is needed to confirm the safety of any change in dose/fractionation in early breast cancer.

Hypofractionation trial results

Hypofractionation entails shortening overall treatment time by increasing the dose per fraction per day. Studied regimens generally reported fractionation schedules of three to four weeks. Phase II trials reported low local failures with acceptable complications, which prompted the initiation of two large international randomised trials testing hypofractionation in early breast cancer.^3,4 Both trials have now reported 10-year follow-up results as well as detailed cosmetic results and 10-year complication (late toxicity) results. 

The largest trial from the UK – the UK Standardisation of Breast Radiotherapy (START) B Trial of radiotherapy hypofractionation for early breast cancer – reported its 10-year results at San Antonio Breast Cancer Symposium in 2012 and published results in The Lancet Oncology in 2013.⁵

A smaller randomised trial from a Canadian trial, led by the Ontario Clinical Oncology Group (OCOG), reported long-term results (median follow-up 12 years) of hypofractionated radiation therapy for breast cancer in the New England Journal of Medicine in 2010.³

START B UK trial

Trial design

Between 1999 and 2001, a total of 2,215 patients were randomised to standard fractionation 50Gy in 25 fractions over five weeks or hypofractionation 40Gy in 15 fractions over three weeks. Eligibility criteria included pathological stage pT1-3 N0-1M0, > 1mm margin of excision, age over 18 years, and no breast reconstruction. The study endpoints were local-regional tumour relapse, defined as reappearance of cancer at irradiated sites, late normal tissue effects, and quality of life. 

Results

At 10-year follow-up, the rate of local regional recurrence (LRC) in the standard fractionation regimen was 5.5% (4.2-7.2) as compared with 4.2% (3.2-5.9) in the hypofractionation arm, demonstrating no significant difference. 

Also at 10 years, there was significantly lower incidence of distant relapse in the hypofractionation arm, which was also noted at five-year follow-up.

Canadian OCOG trial

Trial design 

Between April 1993 and September 1996, a total of 1,234 patients underwent randomisation, with 612 assigned to the control group and 622 to the hypofractionated-radiation group (50Gy in 25 fractions over a period of 35 days versus 42.5Gy in 16 fractions over a period of 22 days). Mean follow-up was 12 years. The exclusion criteria included large breast size, positive surgical margin, and axillary node positive.

The primary end point was local recurrence at 10 years, and secondary end points were a distant (including regional) recurrence of breast cancer; second cancers, including contralateral breast cancer; breast cosmesis; late toxic effects of radiation; and death.

At baseline, 41.8% of patients had received adjuvant tamoxifen, and 10.9% had received adjuvant systemic therapy, most commonly (CMF).

Results

At 10 years, the cumulative incidence of local recurrence was 6.7% in the control group as compared with 6.2% in the hypofractionated-radiation group (absolute difference of 0.5 percentage points; 95% CI, −2.5-3.5). There was no difference in distant free survival or overall survival between the two arms.

Observations

The 10-year results of both trials reporting low local recurrence rates with hypofractionated radiation therapy are reassuring in that the results are consistent in both studies and remain low at long-term follow-up, however, a number of caveats need to be noted before a change in practice recommendation.

Firstly, in both studies, few women received adjuvant systemic chemotherapy and taxanes, and none had received trastuzumab. A total of 22% of patients in the START B trial were treated with chemotherapy, of whom 59·1% received an anthracycline-containing regimen and 39·9% received cyclophosphamide, methotrexate and fluorouracil (CMF) combination therapy alone, while only 10.9% in the OCOG trial had received adjuvant systemic therapy, most commonly CMF. 

In addition, only a small number of patients (79) were treated with nodal radiation in START B trial and none in the OCOG trial.

Boost irradiation of the tumour bed was only used in the Start B Trial. Large breasted patients were excluded from the OCOG trial but not the START B trial.

Cosmesis and toxicity

START B trial cosmesis

Breast cosmesis was prospectively evaluated utilising a subjective self-assessment and objective physician assessment. Photographs were taken at baseline, year one and year five. Results showed that moderate breast shrinkage and breast oedema was lower in the 40Gy group. At year five follow-up significantly better cosmetic results were recorded in this hypofractionated arm. 

This result needs to be confirmed at 10 years although greater than four-year follow-up is deemed adequate to assess cosmetic effect.

START B skin toxicity 

The study found low grade 2 and 4 toxicity based on RTOG-EORTC scale. There was 2.7% and 2.5% grade 2 skin toxicity, respectively, showing no difference between the two arms. Seventy-four percent of patients in the control arm and 64.8% in the hypofractionation arm had an excellent cosmetic score with no difference between the two arms at 10 years.

START B late complications 

Late complications in the START B trial showed that the incidence of rib fracture, lung fibrosis and late cardiac events were all rare and not different in the two arms. Rib fracture was 0.3% in both arms, symptomatic lung fibrosis was 0.2% and ischaemic heart disease was 1.4% and 0.7% (left sided 0.5 and 0.3%) in the control versus hypofractionation arms. Survival was excellent in both arms, with a trend for even greater survival in the hypofractionated arm (84.4% versus 80.4%).

OCOG trial cosmetic score 

The effects of radiation therapy on skin and subcutaneous tissue were graded on a scale of 0-4 (with 0 indicating no toxic effects and grade 4 indicating skin ulceration or soft-tissue necrosis). A trained clinical trials nurse assessed the cosmetic outcome using the EORTC cosmetic rating system. At 10-year follow-up, grade 3 toxicity on skin was 2.7% versus 2.5 %, and 3.6% versus 2.5% on subcutaneous tissues in the standard regimen compared to the hypofractionated regimen. 

OCOG toxicity 

At 10 years, grade 3 radiation associated morbidity was 4% or less in both groups. Cardiac disease was 1.5% in the control group versus 1.9% in the hypofractionated-radiation group, showing no significant difference.

Patient suitability for hypofractionated breast radiation

Based on the results of the two trials, should all categories of patients be recommended for hypofractionated radiation therapy or are there subgroups not included in large numbers in the trial where the data is not robust enough to recommend a change in practice? Subgroups to be considered include young age (< 40 years), large breast size, breast reconstruction, collagen vascular disease, wound complications, chemotherapy, nodal radiation and use of boost.

Young age (< 40 years)

Young age is often reported to be an unfavourable prognostic factor for local control in patients treated with breast conserving therapy (also a factor post mastectomy). In the EORTC 22881-10882 trial,⁶ youth was the single largest significant predictor of local recurrence. A boost is mandatory in this group as confirmed by the EORTC boost trial. 

The Canadian OCOG trial showed that young age was not inferior in the hypofractionated arm, and the START B trial hinted at better local control in the hypofractionated arm. So although a boost was only utilised in START B trial (10Gy in five fractions) there is no evidence that young age should predict fraction size biologically and therefore should not be of concern with hypofractionated radiotherapy.

Large breast size

Large breast size was an exclusion factor in the OCOG trial but not the START B trial. With use of segmented field technique, breast size is only an issue if dosimetry criteria are not met. The size of the hot spot or the heterogeneity of doses within the breast predict for fibrosis and poorer cosmesis than actual breast size. It makes more sense to use dosimetric factors when using segmented planning techniques rather than breast size. As confirmed above, excellent cosmesis can be achieved even with large breast size.

The Intensity Modulated Radiation Therapy (IMRT) approach to breast tangents results in better homogeneity and less acute and late morbidity and is helpful in improving dose homogeneity for large breast size. Therefore, with the utilisation of better quality radiation techniques and optimal planning, large breast size should not be an exclusion criterion for hypofractionated regimen.

 (click to enlarge)

Inclusion of supraclavicular (S/Clav) fossa and axilla

There was no robust data in the randomised trials on the toxicity of S/Clav radiation and hypofractionated regimen as numbers were small in the trial results. There is no level 1 evidence to support hypofractionation in nodal radiation; however, it is in current practice in the UK and Canada. The small number of patients treated with adjuvant nodal radiation in the START B trial did not report any significant toxicity. As a result, this should not be an exclusion criterion for hypofractionated schedules.

Hypofractionation and chemotherapy

START B did include cyclophosphamide, methotrexate and fluorouracil (CMF) and anthrocyclines in 311 patients on the hypofractionated radiation arm while few patients in the OCOG trial were treated with CMF chemotherapy. There was no data on taxanes/trastuzumab interaction with hypofractionated regimens. Radiation should not be delivered concurrently with hypofractionated schedules and care is needed with cardiac doses. 

So, although not tested completely in randomised trials, the addition of systemic chemotherapy should not be an exclusion factor for hypofractionated treatment schedules as long as strict dosimetry targets on cardiac dose are met.

Post-operative wound complications, collagen vascular disease and reconstruction

Seroma is a common complication following breast cancer surgery, resulting in poorer cosmetic result with adjuvant radiation therapy despite the fractionation schedule utilised. Post-operative wound complications also result in poor cosmesis. Longer fractionated schedules could mitigate against late fibrosis, theoretically therefore this would be a more favourable regimen and could reduce ‘hot spots’ in this setting.

Collagen vascular disease, such as systemic lupus erythematosus, has been shown to increase the toxicity of radiation and in some studies increases the fibrosis risk effecting cosmesis and late cosmetic scores. In the past, this was a relative contraindication to a breast conservation approach; however, that is now not the case. More protracted fractionated regimens are generally utilised in this setting. 

Patients with immediate reconstruction should not be treated with hypofractionated regimens, as there is no data on late effects for this group. Patients with significant wound issues post-operatively, including seromas, active collagen vascular disease and immediate reconstruction, should be excluded from hypofractionated regimens. Therefore, these sub-groups are not recommended as suitable for hypofractionated breast radiation.

Current Irish practice 

Arising from the publication of the Canadian OCOG trial’s 10-year results and presentation of the START B trial’s 10-year results at the San Antonio meeting, an audit of all public radiation centres in Ireland was undertaken in February 2013 to assess whether hypofractionated regimens were being recommended for breast cancer patents. The results showed consistently across all sites that only 25% (range 14-37%) of patients were treated with a hypofractionated regimen, while conventional fractionation (using a 5-6.5 week regimen) was found to be the favoured radiation schedule. 

Admittedly, this survey predates the final publication of the START B trial and also prior to this topic being reviewed at the Annual National Radiation Oncology meeting in Cork University Hospital (April 2013), at which the author delivered a lecture titled Optimum fractionation schedule for early breast cancer.

In contrast, a similar audit was undertaken in 2011 across all radiation centres in British Columbia in Canada. The standard hypofractionation schedule, as per the OCOG trial, is a 16-fraction regimen and the audit reported that over 80% of patients with early breast cancer were treated with this regimen. The results were similar across all radiation centres.

This discrepancy in adopting the new regimen in Ireland will be explored in a future audit, which will take a more in-depth look at three difference time lines, spanning 2010 to 2014, to assess if the published START B results influenced the use of hypofractionation for early breast cancer and ascertain any barriers to practice change in breast radiation practice in Ireland.

Conclusion

Hypofractionated radiation therapy, utilising a three-week regimen (plus boost in selected patients), provides excellent local control with recurrence rates equivalent to more protracted treatment regimens, and has the secondary benefit of improved toxicity and cosmesis. Hypofractionation should be the standard treatment approach for the majority of early breast cancer patients.

References

1. Clarke M, Collins R, Darby S, et al. (Early Breast Cancer Trialists’ Collaborative Group (EBCTCG)). Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials. Lancet 2005; 366: 2087-2106. 
2. Bartelink H, Horiot JC, Poortmans PM, et al. Impact of a higher radiation dose on local control and survival in breast-conserving therapy of early breast cancer: 10-year results of the randomized boost versus no boost EORTC 22881-10882 trial. J Clin Oncol 2007; 25: 3259-3265.
3. Whelan TJ, Pignol J-P, Levine MN, et al. Long-term results of hypofractionated radiation therapy for breast cancer. NEJM 2010; 362: 513-520.
4. Bentzen SM, Agrawal RK, Aird EG, et al. (START Trialists’ Group). The UK Standardisation of Breast Radiotherapy (START) Trial B of radiotherapy hypofractionation for treatment of early breast cancer: a randomised trial. Lancet 2008; 371(9618): 1098-1107.
5. Haviland JS, Owen JR, Dewar JA, et al. (START Trialists Group). The UK Standardisation of Breast Radiotherapy (START) trials of radiotherapy hypofractionation for treatment of early breast cancer: 10-year follow-up results of two randomised controlled trials. Lancet Oncol 2013; 14(11): 1086-1094.
6. Bartelink H, Horiot JC, Poortmans PM, et al. Impact of a higher radiation dose on local control and survival in breast-conserving therapy of early breast cancer: 10-year results of the randomized boost versus no boost EORTC 22881-10882 trial. J Clin Oncol 2007; 25(22): 3259-65.