Long term survival and local control outcomes from single dose targeted intraoperative radiotherapy during lumpectomy (TARGIT-IORT) for early breast cancer: TARGIT-A randomised clinical trial
BMJ 2020; 370 doi: https://doi.org/10.1136/bmj.m2836 (Published 19 August 2020) Cite this as: BMJ 2020;370:m2836Linked Opinion
Targeted intraoperative radiotherapy for early breast cancer—new evidence challenges traditional treatment
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Dear Editor,
We would also like to reassure BMJ readers that the original power calculations were performed for a trial in which randomisation occurs before surgical excision of the cancer, and compared the standard conventional fractionated whole breast external beam radiotherapy (EBRT) vs. risk-adapted single dose targeted intraoperative radiotherapy during lumpectomy (TARGIT-IORT). The sample size planned as per these calculations was 2232, and the trial in fact recruited 2298 patients. So, this is a very adequate sample size, and as shown in the original Figure 6[1], is larger than other trials of partial breast irradiation for invasive breast cancer.
As mentioned in our previously published papers, an additional and separate cohort of patients was added 4 years after commencement of the trial[1-3]. Patients for this separate cohort were treated completely differently – they were recruited after they already had the surgical excision of their cancer, with full histopathology review. It was more restrictive than the main trial, i.e, the characteristics of the tumours in this separate trial was different. They were then randomised using a separate randomisation table, to either ‘delayed single-dose radiotherapy’, or ‘whole breast external beam radiotherapy’. When randomised to ‘delayed single-dose radiotherapy’, it was undertaken through a second operation, by reopening the lumpectomy wound and inserting the radiotherapy applicator in the reopened wound. This was done at a median of 37 days of delay after the initial surgical excision. The control arm received standard post-operative whole breast external beam radiotherapy (no re-opening of the wound).
While these two cohorts were previously published within the same manuscript, they were always regarded as separate studies, but simply nested together. It is important to recognise that they had different eligibility criteria and different experimental treatments. Of course, they had separate randomisation tables, such that the randomisation for one trial was immediate TARGIT-IORT vs EBRT (n=2298), and for the other trial, delayed TARGIT-IORT vs EBRT (n=1153). A planned separate analysis of these two studies was specified from the outset. Before the data were unblinded for the current analysis, i.e., before any outcome data had been seen by anyone, the Independent Trial Steering Committee recommended that they should be published separately. This is what we have done[1-3].
However, wanting to respond to Yarnold et al and although not specified in the statistical analysis plan, we sought the advice of the TSC who agreed that we could present the following limited analysis. The analysis for non-inferiority, as suggested by Yarnold and colleagues’ 4th reference, shows that when the two trials are combined, the Kaplan-Meier estimate of local recurrence at 5-year complete follow up for the two experimental arms together (immediate TARGIT-IORT and delayed TARGIT-IORT, taken together) was 2.86% (SE 0.41). For the two control arms together, it was 1.04% (SE 0.25). The difference was 1.82%, with an upper 90% CI of 2.49%, which does not cross 2.5%. Thus, non-inferiority of TARGIT-IORT for local recurrence was confirmed for the two trials combined. Furthermore, there was no difference in breast cancer mortality - HR 1.14 (95% CI 0.84 – 1.56), and a reduction in non-breast-cancer mortality - HR 0.74 (95%CI 0.56 – 0.97) favouring the TARGIT-IORT arms.
Even if this combined analysis of these two trials had not confirmed non-inferiority, the results of individual separate trials would still remain the valid primary outcomes for informing clinical practice. It is important to recognise this point, with the perspective that TARGIT-A is a non-inferiority trial, which is quite distinct from a superiority trial. The basic premise is not to prove superiority of the new treatment, but in a setting where this new treatment carries obvious advantages, a non-inferiority margin is selected at a level which would be clinically relevant to patients. The TARGIT-A trial clearly fits the bill in all respects, and our decision to use 2.5% as the non-inferiority margin is in fact a stringent one. The obvious benefits to the patient with TARGIT-IORT is of completing the radiation treatment during surgery in 8 out of 10 patients, in addition to the many other benefits specified above.
The long-term comparable effectiveness of risk-adapted single-dose TARGIT-IORT during lumpectomy to EBRT was confirmed, with equivalent local and distant control, breast preservation, and breast cancer mortality, along with a substantial reduction in non-breast cancer mortality. For the long-term outcomes of delayed second-procedure TARGIT-IORT vs EBRT, the difference in 5-year local recurrence rate was above the noninferiority margin, but no statistically significant difference was found in local and distant control, breast preservation and mortality. These separate analyses of each of the trial remain the agreed primary analyses and outcomes[1-3].
In fact, all these results[1-3] are consistent with our previous 2014 publication. The BMJ paper, and incidentally, in respect of the delayed second-procedure placement of the radiotherapy device, the one recently published in JAMA Oncology[8,10], both now supersede it, because of the accrual of more data and a longer follow up.
What is clear is that many other countries across the world have now enthusiastically embraced this approach, even offering immediate intra-operative radiotherapy as part of national guidelines or consensus statements. Over 45,000 patients have now had this treatment at 260 centres in 38 countries, with the many benefits we outline previously[4-10]. Have they all got it wrong? In the UK, where the trial was designed and co-ordinated, NICE gave its recommendation in January 2018 for the treatment to be made available at centres where the equipment and expertise was already available, with the stipulation that further data accrual should continue. With the BMJ publication now offering further solid proof of benefit and an excellent level of follow-up, it is now time to roll this out further.
Finally, as we have stated before, the rulings of the UK Supreme Court (Montgomery v Lanarkshire Health Board, 2015), the Royal College of Surgeons, (https://www.rcseng.ac.uk/standards-and-research/standards-and-guidance/g...), and the General Medical Council (https://www.gmc-uk.org/ethical-guidance/ethical-guidance-for-doctors/con...) make it crystal-clear that patients have the right and doctors have the duty to discuss the full range of possible treatments, not simply limited to the simple preference of the doctor who is advising them.
References
1. Vaidya, J.S., et al., Effect of Delayed Targeted Intraoperative Radiotherapy vs Whole-Breast Radiotherapy on Local Recurrence and Survival: Long-term Results From the TARGIT-A Randomized Clinical Trial in Early Breast Cancer. JAMA Oncol, 2020. 6(7): p. e200249.
2. Vaidya, J.S., et al., Long term survival and local control outcomes from single dose targeted intraoperative radiotherapy during lumpectomy (TARGIT-IORT) for early breast cancer: TARGIT-A randomised clinical trial. BMJ, 2020. 370: p. m2836.
3. Vaidya, J.S., M. Bulsara, and M. Baum, Targeted Intraoperative Radiotherapy for Early Breast Cancer-Reply. JAMA Oncol, 2020.
4. Coombs, N.J., et al., Environmental and social benefits of the targeted intraoperative radiotherapy for breast cancer: data from UK TARGIT-A trial centres and two UK NHS hospitals offering TARGIT IORT. BMJ Open, 2016. 6(5): p. e010703.
5. Sperk, E., et al., Late radiation toxicity after intraoperative radiotherapy (IORT) for breast cancer: results from the randomized phase III trial TARGIT A. Breast Cancer Res Treat, 2012. 135(1): p. 253-60.
6. Welzel, G., et al., Radiation-related quality of life parameters after targeted intraoperative radiotherapy versus whole breast radiotherapy in patients with breast cancer: results from the randomized phase III trial TARGIT-A. Radiat Oncol, 2013. 8(1): p. 9.
7. Andersen, K.G., et al., Persistent pain after targeted intraoperative radiotherapy (TARGIT) or external breast radiotherapy for breast cancer: A randomized trial. Breast, 2012. 21(1): p. 46-49.
8. Keshtgar, M.R., et al., Objective assessment of cosmetic outcome after targeted intraoperative radiotherapy in breast cancer: results from a randomised controlled trial. Breast Cancer Res Treat, 2013. 140(3): p. 519-25.
9. Corica, T., et al., Cosmesis and Breast-Related Quality of Life Outcomes After Intraoperative Radiation Therapy for Early Breast Cancer: A Substudy of the TARGIT-A Trial. Int J Radiat Oncol Biol Phys, 2016. 96(1): p. 55-64.
10. Corica, T., et al., Cosmetic outcome as rated by patients, doctors, nurses and BCCT.core software assessed over 5 years in a subset of patients in the TARGIT-A Trial. Radiat Oncol, 2018. 13(1): p. 68.
Competing interests: As in the main paper
Dear Editor,
We were pleased to see the rapid response from Dr Singer, Dr DeSilva-Minor and Prof Bleddyn Jones. As they accurately point out, our paper shows that for every 8 out of every 10 patients receiving TARGIT-IORT, no further radiotherapy is required. That means no unnecessary hospital visits, no unnecessary over-treatment, no unnecessary exposure to a potentially lethal virus and many other long-term benefits as we have outlined in a series of publications relating to cosmesis, quality of life, breast pain and so on[1-7]. As practicing clinicians, they recognise and underscore the importance of avoiding the scattered radiation that inevitably accompanies whole breast radiotherapy but is avoided by TARGIT-IORT.
Turning to the first part of the response from Yarnold et al, we would like to reassure them that the TARGIT-A trial has consistently had the highest level of governance. In addition to the international steering committee (ISC), a core group managed the trial within the UCL trials unit. The ISC needed to have the manufacturer as observer members, and this was acknowledged and required by the sponsor. As per advice from the UCL sponsoring office, their presence was essential for assessment of safety of what was then a new device (albeit CE marked), and for helping us to identify and initiating new centres for recruitment of patients in the trial. Furthermore, independent oversight was provided by the DMC. This arrangement was approved by of the HTA/NIHR, the funders of the trial. After all patients had been recruited, received their treatment and the first analysis were presented, the DMC, having completed their remit, were thanked and disbanded. Trial governance was then handed over to the HTA/NIHR-appointed independent Trial Steering Committee (TSC). As has been stated in all related records and publications, good clinical practice principles were followed throughout.
Further queries are dealt with in our next response.
References
1. Coombs, N.J., et al., Environmental and social benefits of the targeted intraoperative radiotherapy for breast cancer: data from UK TARGIT-A trial centres and two UK NHS hospitals offering TARGIT IORT. BMJ Open, 2016. 6(5): p. e010703.
2. Sperk, E., et al., Late radiation toxicity after intraoperative radiotherapy (IORT) for breast cancer: results from the randomized phase III trial TARGIT A. Breast Cancer Res Treat, 2012. 135(1): p. 253-60.
3. Welzel, G., et al., Radiation-related quality of life parameters after targeted intraoperative radiotherapy versus whole breast radiotherapy in patients with breast cancer: results from the randomized phase III trial TARGIT-A. Radiat Oncol, 2013. 8(1): p. 9.
4. Andersen, K.G., et al., Persistent pain after targeted intraoperative radiotherapy (TARGIT) or external breast radiotherapy for breast cancer: A randomized trial. Breast, 2012. 21(1): p. 46-49.
5. Keshtgar, M.R., et al., Objective assessment of cosmetic outcome after targeted intraoperative radiotherapy in breast cancer: results from a randomised controlled trial. Breast Cancer Res Treat, 2013. 140(3): p. 519-25.
6. Corica, T., et al., Cosmesis and Breast-Related Quality of Life Outcomes After Intraoperative Radiation Therapy for Early Breast Cancer: A Substudy of the TARGIT-A Trial. Int J Radiat Oncol Biol Phys, 2016. 96(1): p. 55-64.
7. Corica, T., et al., Cosmetic outcome as rated by patients, doctors, nurses and BCCT.core software assessed over 5 years in a subset of patients in the TARGIT-A Trial. Radiat Oncol, 2018. 13(1): p. 68.
Competing interests: As in the main paper
Dear Editor
The TARGIT-A trial was conducted outside the research governance framework described in UK Medical Research Council Guidelines for Good Clinical Practice in Clinical Trials, published in 1998 (1,2). The Guidelines explain how individuals responsible for initiating and supervising the day-to-day running of trials constitute a Trial Management Group (TMG). TMGs report at least annually to a small independent Trial Steering Committee (TSC) of experts appointed by the research sponsor (in this case, University College London) that takes legal responsibility for the scientific and ethical integrity of research. In addition, an independent Data Monitoring Committee (DMC) reports to the TSC. The DMC is the only body authorised to review interim analyses prepared in confidence by the trial statistician. Its responsibility is to review data quality and to monitor trial progress against pre-specified criteria.
Why is this relevant? The first publication of trial outcomes in 2010 lists an “International Trial Steering Committee" comprising 22 individuals, all closely involved in the trial, including a past chairman drawing monthly consultancy fees from Zeiss and several Zeiss employees, so neither small nor independent (3). This committee is actually the TMG as per MRC definitions. For the first 14 years of the trial, there was no independent TSC to which the TARGIT-A trial DMC could report. This weakness raised problems for the DMC chair, Professor J. Cuzick, and ended with the TMG disbanding its only oversight committee (4).
The 2020 publications of pre- and post-pathology strata as independent trials represent marked departures from previous publications and serious deviations from standard practice (6,7). The first report of trial outcome in 2010 introduces two strata, pre- and post-pathology, within the context of a single trial as illustrated in Figure 2 (3). The statistical analysis plan is appropriately applied to the whole trial population of 2232 patients with no reference to strata. Supplementary analyses of strata first appear in the 2014 publication, where they are introduced as ‘firmly rooted in the principles of pragmatism’ (5). The pragmatism is fine but is followed by retrospective theoretical justifications for testing greater effectiveness of both pre- and post-pathology approaches. This creates three non-inferiority tests out of one, each with a chance of meeting the pre-defined margin for non-inferiority. The recent publication restricted to the supplementary pre-pathology stratum states that the statistical analysis plan for TARGIT-A was modified before unblinding of the data, but whether this modification was introduced before unblinding in/before 2013 or 2019 is not clear (7). Even with effective blinding, failure to meet the non-inferiority criterion will have been evident to the data analytical team in advance of any decision to publish the strata as separate trials.
The appointment of a truly independent TSC is welcome, and credit is due to the UK Health Technology Agency (HTA) for taking this essential initiative. It comes 15 years after trial activation but not too late to insist on publication of the planned analysis of the whole trial population. The 2016 TARGIT-A trialists’ report to the UK Health Technology Agency (HTA) re-confirms a trial sample size calculated for local recurrence within the whole patient population and a main analysis planned to be performed on the whole population (8). The latter should be published without further delay with the supplementary analyses of strata presented as hypothesis-generating and inadmissible as evidence of clinical benefit.
Yours faithfully
Professor John R Yarnold
Emeritus Professor of Clinical Oncology
The Institute of Cancer Research, London, UK
Dr Anna Kirby
Consultant in Clinical Oncology
The Royal Marsden NHS Foundation Trust & The Institute of Cancer Research, London, UK
Professor Soren M Bentzen
Department of Epidemiology & Public Health, University of Maryland School of Medicine
Baltimore, MD, U.S.A.
1. MRC Guidance on Good Clinical Practice in Clinical Trials (1998) http://mrc.ukri.org/documents/pdf/good-clinical-practice-in-clinical-trials
2. Yarnold J. [ Start of cheaper techniques for breast cancer delayed in UK despite adoption elsewhere: Response ] https://www.bmj.com/content/350/bmj.h2874/rapid-responses.
3. Vaidya JS, Joseph DJ, Tobias JS, Bulsara M, Wenz F, Saunders C, et al. Targeted intraoperative radiotherapy versus whole breast radiotherapy for breast cancer (TARGIT-A trial): an international, prospective, randomised, non-inferiority phase 3 trial. Lancet. 2010;376(9735):91-102.
4. Cuzick J. Radiotherapy for breast cancer, the TARGIT-A trial. Lancet. 2014 May 17;383(9930):1716.
5. Vaidya JS, Wenz F, Bulsara M, Tobias JS, Joseph DJ, Keshtgar M, et al. Risk-adapted targeted intraoperative radiotherapy versus whole-breast radiotherapy for breast cancer: 5-year results for local control and overall survival from the TARGIT-A randomised trial. Lancet. 2014;383(9917):603-13.
6. Vaidya JS, Bulsara M, Saunders C, Flyger H, Tobias JS, Corica T, et al. Effect of Delayed Targeted Intraoperative Radiotherapy vs Whole-Breast Radiotherapy on Local Recurrence and Survival: Long-term Results From the TARGIT-A Randomized Clinical Trial in Early Breast Cancer. JAMA Oncol. 2020:e200249.
7. Vaidya JS, Bulsara M, Baum M, Wenz F, Massarut S, Pigorsch S, et al. Long term survival and local control outcomes from single dose targeted intraoperative radiotherapy during lumpectomy (TARGIT-IORT) for early breast cancer: TARGIT-A randomised clinical trial. BMJ. 2020;370:m2836.
8. Vaidya JS, Wenz F, Bulsara M, Tobias JS, Joseph DJ, Saunders C, et al . An international randomised controlled trial to compare TARGeted Intraoperative radioTherapy (TARGIT) with conventional postoperative radiotherapy after breast-conserving surgery for women with early-stage breast cancer (the TARGIT-A trial). Health Technol Assess 2016;20 (73).
Competing interests: No competing interests
Dear Editor,
We read with interest the BMJ article on long term follow up data for Intraoperative radiotherapy for early stage breast cancer.
We wish to make the following points:
Intraoperative breast radiation therapy is not new and has been investigated by many groups in the past with varying degrees of success. Many of these studies were perhaps neglected because they were small, often non-randomised, did not stratify patients adequately, and were carried out many years ago [1].
Systemic treatments were less effective in the 1990s. There was less knowledge about biological subtypes of breast cancer. Human Epidermal Receptor (HER2) targeted treatment did not exist. Chemotherapeutic agents such as Taxanes, and endocrine endocrine therapies such as the aromatase inhibitors were not in use. Improvements in systemic therapies have been shown not only to reduce distant metastases but also improve local control.
Prime II and CALGB 9343 studies looking at breast radiation versus no radiation recruited lower risk groups: Prime II: Women 65 years and over, grade I or II node negative, tumours less than 3 cm; CALGB 9343 70 years and over, T1N0, ER positives only[2,3]. The TARGIT A study recruited women over age 45, all grades and a significant proportion (20%) were 1 - 4 node positive and/or grade III. These were different populations and not directly comparable.
There is data to show that whole breast radiotherapy may lead to cardiac and lung long term damage and second malignancies, especially in women who smoke over the subsequent 10 to 20 years [4]. More women are surviving long term. There is an increased incidence in primary lung cancers but also other malignancies including oesophageal cancer. Deep inspiratory breath hold techniques for left breast radiotherapy is now standard practice to greatly reduce the radiation dose to the heart. Whilst this may protect the heart against damage to the left anterior descending coronary artery, there is no evidence that this will reduce the amount of radiation to the lung. Intraoperative radiotherapy with the Intrabeam machine would greatly reduce radiation to both heart and lung tissue.
The TARGIT A study, Intrabeam machine delivers 50 Kv X rays dose of 20 Gy to the surface of the applicator, 6-7 Gy at 1 cm depth and 2-3 Gy at 2 cm depth depending on the size of applicator. As expected with brachytherapy, there is a rapid fall off in radiation dose. For the elimination of a small rim of cancer cells beyond the excision, the high dose close to the surface of the applicator with low energy X rays (which are more biologically efficient) may be more effective than uniform dose from an external beam radiotherapy, partly compensating for the lower doses at larger distances from the tumour bed. Water phantom and cell line studies indicate that 50 Kv x rays are of different quality than 6 mv photons, resulting in a higher relative biological effect (RBE) range 1.2 – 1.5 [5]. The potential advantages in the delivery of a single high dose radiation immediately at tumour resection, are in the elimination of repopulation time for any tumour cells left in the tumour bed; reduced double strand DNA break repair due to high dose as shown in cell line studies and possible indirect effect on vasculature, stroma and of immune system [6]. The relatively small volume of normal tissue irradiated should result in a reduction in normal tissue side effects.
In the current Covid Pandemic, many of the UK radiotherapy centres are now adopting the FAST FORWARD hypofractionated regime 26 Gy in 5 daily fractions to the whole breast [7]. This would involve a number of visits to the radiotherapy centre. There is also great interest in partial breast irradiation the IMPORT LOW trial using 15 fractions to a higher dose [8]. Whether 5 daily fractions or 15, this still involves significant patient travel to the radiotherapy centre. Often patients will recruit family members to help with car transport and hospital parking. This will have socio-economic impact and more recently, there would be a risk of Covid infection. This risk would be eliminated with Intrabeam intraoperative radiotherapy.
Over the last 30 years breast cancer local control has improved greatly. Five year local relapse rates for lower risk groups are now in the region of 1 – 2 %. Previous data suggested that if a woman suffered a local relapse, her overall survival was significantly worse. [9]. This situation has changed. The TARGIT A trial at 5 years showed 0.95% breast relapse for whole breast radiotherapy versus 2.11% for IORT risk adapted approach. However, there was a non-significant trend toward an improvement in overall survival for the IORT arm.
This technique should be approved fully by NICE and funding made available by NHS England, particularly in those centres which have the equipment and expertise. There should be an option for patients to discuss the risk / benefits for this treatment.
References:
1 Polgar C et al. Patient selection for accelerated partial-breast irradiation (APBI) after breast-conserving surgery: Recommendations of the Groupe Européen de Curiethérapie-European Society for Therapeutic Radiology and Oncology (GEC-ESTRO) breast cancer working group based on clinical evidence (2009) Radiotherapy and Oncology. 2010: 94: 264-273.
2 Kunkler IH et al. Breast-conserving surgery with or without irradiation in women aged 65 years or older with early breast cancer (PRIME II): a randomised controlled trial. Lancet Oncology. 2015: 16 (3) 266-273.
3 Hughes KS et al. Lumpectomy plus tamoxifen with or without irradiation in women age 70 years or older with early breast cancer: long-term follow-up of CALGB 9343. 2013. J Clin Oncol.(19): 2382-2387.
4 Taylor C et al. Estimating the Risks of Breast Cancer Radiotherapy: Evidence From Modern Radiation Doses to the Lungs and Heart and From Previous Randomized Trials. J Clin Oncol. 2017. (35); 1641-1649.
5 Liu Q et al. Relative Biologic Effectiveness (RBE) of 50 kV X-rays measured in a phantom for intraoperative tumor-bed irradiation. 2013. Int J Radiat Oncol Biol Phys. 85(4); 1127-33.
6 Herskind C et al. Biology of high single doses of IORT: RBE, 5 R’s, and other biological aspects. Radiation Oncology. 2017; 12:24 : 1-14
7 Brunt AM et al. Hypofractionated breast radiotherapy for 1 week versus 3 weeks (FAST-Forward): 5-year efficacy and late normal tissue effects results from a multicentre, non-inferiority, randomised, phase 3 trial. 2020. Lancet. 395; 1613-26
8 Coles CE et al. Partial-breast radiotherapy after breast conservation surgery for patients with early breast cancer (UK IMPORT LOW trial): 5-year results from a multicentre, randomised, controlled, phase 3, non-inferiority trial. 2017. Lancet. 390: 1048-1060.
9 Wapnir IL, et al. Prognosis after ipsilateral breast tumor recurrence and locoregional recurrences in five National Surgical Adjuvant Breast and Bowel Project node-positive adjuvant breast cancer trials. J Clin Oncol. 2006;24:2028–37
Competing interests: No competing interests
Dear Editor,
With respect, Mrs Hazel Thornton is incorrect in relation to the chairmanship of the steering committee (Rapid Response posted on 17 Sep 2020).
In 2013, Prof Michael Baum (UK) and Prof David Joseph (Australia), as joint chairs of the International Steering Committee (ISC), handed over the governance of this trial to the Trial Steering committee (TSC), with Prof Freddie Hamdy (University of Oxford) as its independent chair. Prof Jayant Vaidya has never been the chair of the ISC, and the ISC did not co-exist together with the TSC.
The TSC was independently appointed by the National Institute for Health Research (NIHR) who followed their rule that at least 75% of the membership should be independent, and appointed just two members of the project team - the chief investigator (Prof Jayant Vaidya) and the trial statistician (Prof Max Bulsara). As stated in the paper, this TSC has provided oversight for the trial since the publication of the last results.
Competing interests: As in the main paper
Dear Editor,
It is good that Targit-A trial investigators agreed, post hoc, the necessity of explicitly acknowledging the work of Data Monitoring Committees (DMC) in trial reports, following a challenge to this omission. [1] (Their rapid response 4th September 2020.) Achieving transparency about all independent oversight of trial data is essential in trial reports.
Oversight of the Targit-A trial was undertaken by the DMC from its` appointment in July 2006 until it was disbanded 24th May 2013. Trial recruitment was completed 25th June 2012. Publication of the initial 5-year results occurred in June 2013. [2] At this time the chairmanship of the Trial Steering Committee also changed: Jayant Vaidya took over from Michael Baum and David Joseph. The current report [3] is of the follow-up findings of the Targit-A trial.
We learn from this long-term follow-up that “Since the last analysis, the trial oversight has been provided by an independent steering committee, appointed by the Health Technology Assessment (HTA) programme of the National Institute of Health Research, Department of Health and Social Care, UK, which also includes a patient as a member”. Authors also provide name and institution of individual Independent Steering Committee (ISC) members “for providing trial oversight” in the paragraph of thanks at the end of their paper. If I have understood correctly, this committee is separate and different from the Trial Steering Committee (TSC), except for two non-independent members, Jayant Vaidya and Max Bulsara, who were common to both ISC and TSC. Bearing this in mind, how was independence maintained?
Also, how does `oversight by an independent steering committee` differ from oversight provided by a DMC? Did ISC members sign up to a remit document for that trial? The work of DMCs has been studied, defined and made publicly available, enabling researchers to adhere to formulated guidelines for improving the quality of research. [4] [5] The Equator Network https://www.equator-network.orgreporting-guidelines/ provide reporting guidelines about DMCs.
Enlightenment would be welcome.
[1] Hazel Thornton. BMJ rapid response https://www.bmj.com/content/370/bmj.m2836/rr-1 26th August 2020
[2] Vaidya JS, Wenz F, Bulsara M, et al., TARGIT trialists’ group. Risk-adapted targeted intraoperative radiotherapy versus whole-breast radiotherapy for breast cancer: 5-year results for local control and overall survival from the TARGIT-A randomised trial. Lancet2014;383:603-13. doi:10.1016/S0140-6736(13)61950-9. pmid:24224997
[3] Jayant S. Vaidya, Max Bulsara, Michael Baum, Frederik Wenz, Samuele Massarut, et al. Long term survival and local control outcomes from single dose targeted intraoperative radiotherapy during lumpectomy (TARGIT-IORT) for early breast cancer: TARGIT-A randomised clinical trial. BMJ 2020;370:m2836
[4] Jennifer S. Gewinder, Rachel A Kitt, Matthew R. Hunsinger, Joseph Poku, Jacqueline Lozano et al. Reporting of data monitoring boards in publications of randomised clinical trial is often deficient: ACTION systematic review. DOI: 10.1016/j.jclinepi.2016.12,018. 2017.
[5] Damocles Study Group*, a proposed charter for clinical trial data monitoring committees: helping them to do their job well. The Lancet. February 19th 2005. DOI: https://doi.org/10.1016/S0140-6736(05)17965-3
*AM Grant, DG Altman, AG Babiker, MK Campbell, et al.
Competing interests: No competing interests
Dear Editor,
As an `Independent Citizen Advocate for Quality in Research and Healthcare` I thank Jayant Vaidya and his team for their kind words.[1] As `an independent citizen advocate`, it is important for me to emphasise two things.
Firstly, I would ask them to appreciate the difference between being an `Independent Citizen Advocate` (as I described myself in my original rapid response) and an `independent patient advocate` as described in their response. They rightly refer to my campaign `to make clinical evidence easily understood by patients`, but I have no mandate to claim to advocate for patients. I advocate for `quality in research and healthcare`, in which attention to detail is all important.
Which brings me to my second point. I was pleased to see that they recognise `the great importance of the independent data monitoring committee`. Linking these two points, in my rapid response I referred to two solid pieces of work, [2] [3] one a systematic review, concerning the necessity for acknowledging the work of Data Monitoring Committees in reports of trials, saying that I believed it to be a serious omission on their part not to have done so. Detailing the work of a DMC in a trial report improves the quality of that report and ultimately the quality of research – my stated aim. Nearly a decade ago, I publicly advocated that `Better reporting of better research = better healthcare`. [4] It was thus the contribution of the Data Monitoring Committee, as described in the guidance in the systematic review, and the DAMOCLES Group [2][3], that I had hoped to see acknowledged in their report. A `serious omission` not to have done so.
Readers of trial reports, and all those affected by its content, need to know whether or not a trial has had the benefit of the oversight of an independent DMC for the sake of clarity and confidence. Attention to this particular detail is of considerable importance for everyone.
[1] Jayant S. Vaidya, Max Bulsara, Michael Baum, Frederik Wenz, Samuele Massarut, et al. Long term survival and local control outcomes from single dose targeted intraoperative radiotherapy during lumpectomy (TARGIT-IORT) for early breast cancer: TARGIT-A randomised clinical trial. BMJ 2020;370:m2836
[2] Jennifer S. Gewandter, Rachel A Kitt, Matthew R. Hunsinger, Joseph Poku, Jacqueline Lozano et al. Reporting of data monitoring boards in publications of randomised clinical trial is often deficient: ACTION systematic review. DOI: 10.1016/j.jclinepi.2016.12,018. 2017.
[3] Damocles Study Group*, a proposed charter for clinical trial data monitoring committees: helping them to do their job well. The Lancet. February 19th 2005. DOI: https://doi.org/10.1016/S0140-6736(05)17965-3
*AM Grant, DG Altman, AG Babiker, MK Campbell, et al.
[4] Thornton H. Better reporting of better research = better healthcare. EQUATOR NETWORK 3RD Annual Lecture. October 2011. https://www.equator-network.org/2011/10/03/better-reporting-of-better-re...
Competing interests: No competing interests
Dear Editor,
We wish to address the unjustifiable criticisms from Kirby et al. It would take up too much room to address every one of the issues, so we will focus on the most inaccurate and misleading of the points that they make.
They seem to be recommending the Fast-Forward regimen[1], as a type of partial breast irradiation, whereas in reality it is whole breast radiotherapy. The statement “..this partial breast radiotherapy will now be delivered in 5 treatments over one week (their ref 12 cites the Fast Forward trial[1])”, seems to refer to an imaginary and untested combination of the IMPORT-Low (partial breast irradiation that would need about 17 hospital visits and Fast-Forward regimen (whole breast radiotherapy that would still need 7 to 15 hospital visits). Their contention that partial breast irradiation as per the IMPORT-Low protocol is now standard of care is also questionable.
In describing their cosmetic outcomes from the Fast Forward trial, they fail to mention the significant toxicity experienced by patients in with this regimen. In the Fast-Forward trial[1], a substantial proportion of patients report moderate/marked toxicity. For example, 25% of patients receiving the 5-day regimen complained of a ‘harder or firmer breast’. This is consistent with their clinician-assessments of moderate to marked induration (firmness) of the breast outside the tumour area (incidence ratio of 19.1 (95%CI 2.57, 141.9) , p<0.0001 Supplemental Table A2 from reference[1]). This may be a 19-fold increase in only a small incidence by comparison with the control arm (1 vs 20 patients), but it is hardly an insignificant result. It means that a substantial portion of the breast was reportedly 19 times more likely to have moderate or marked induration compared with the control arm. We would like to remind those less familiar with radiotherapy toxicity that firmness and hardness of the breast are likely to be irreversible. We therefore find it surprising that with this high level of patient-reported toxicity, the authors claim that this intense and compressed regimen, although effective, is also “as safe in terms of normal tissue effects”.
Furthermore, this regimen still irradiates the whole breast with its well-known hazards. No reduction in mortality was observed in the trial, and Fast-Forward still requires 7 to 15 hospital visits, including the consultation, planning, daily radiotherapy, and boost, when indicated. In addition, these patients face the inevitable delays between referral for radiotherapy, further referral for radiotherapy planning and an additional interval before the treatment itself actually starts. We are concerned that only this regimen (or no radiotherapy at all) might be offered to patients during the lengthy COVID-19 pandemic. In contrast, partial breast irradiation reduces non-breast cancer mortality and overall mortality[2, 3], and when given intraoperatively as TARGIT-IORT, the many additional advantages include lower toxicity[4], better quality of life[5, 6], and the avoidance of multiple hospital visits for post-operative radiotherapy[7]. There is even lower toxicity in the group receiving TARGIT-IORT alone, for all toxicities (pain, hyperpigmentation, lymphedema of the arm, ulceration, retraction, breast oedema, telangiectasis and fibrosis). Our group have previously confirmed that patients receiving TARGIT-IORT alone are at half the risk of developing higher grade toxicity (HR 0.46; p=0.010)[4]. Furthermore, patients who receive supplemental EBRT are not at a higher risk of developing this type of toxicity, with long-term results up to 12 years[8].
They claim that ‘no radiotherapy’ might achieve the same outcomes as we have achieved in the TARGIT-A trial, without reference to three important published trials[9-11] that refute this assertion. As seen in the table (click the link to see table), patients in the trials comparing radiotherapy vs no-radiotherapy (CALBG, BASO-II and PRIME-II), were highly selected for their low recurrence risk. They were limited to those older than 65 or 70, with smaller, lower grade, node negative and ER positive tumours. This was in contrast with the TARGIT-A trial, which did not have these eligibility limitations, and included a significant number of patients at a much higher risk of local recurrence. Despite this, the 5-year local recurrence rates in trials of ,no-radiotherapy were 2 to 3 times higher than that seen with TARGIT-IORT (see table https://bit.ly/352bjRm)
Most surprisingly, they state that many patients were lost to follow up when in reality, as stated in the paper, 95% had complete follow up for assessment of the primary outcome at 5 years. Furthermore, the observed follow up closely matched the expected follow up, as shown in figure 2 of the paper.
This comment is particularly interesting because in their own Fast Forward paper they state that “5-year visit forms were available for 3681 (96%) patients of 3833 still in follow-up (not died, withdrawn, or lost).” This could not have been strictly true because at this point in follow up, the number ‘at risk’ on the Kaplan-Meier plot of overall survival at the 5-year mark (adding all 3 treatment groups together) was down to just 3213 patients, with 657 patients censored. These 657 censored patients would of course have been seen on or before the 5th anniversary of their day of randomisation. Therefore, unless 468 (3682 minus 3213) of the 657 patients were genuinely seen at the 5th anniversary of their day of randomisation, the completeness of follow up at 5 years would actually be lower than the 96% that they have claimed. If all these patients were seen before the 5th anniversary then the figure would be actually 3213/3833 which is 83.8%, rather than the 96% that they have led the readers to believe.
They claim that there was no external governance in the TARGIT-A trial, when it is explicitly stated in the paper :“Since the last analysis, the trial oversight has been provided by an independent steering committee, appointed by the Health Technology Assessment (HTA) programme of the National Institute of Health Research, Department of Health and Social Care, UK, which also includes a patient as a member”.
They imply that the difference in non-breast cancer mortality was most likely due to imbalance of medical problems and smoking history between the two arms of the trial. However, that is highly improbable in such a large randomised trial. After all, the strength of any RCT is that it avoids imbalances in risk factors both known and unknown. Certainly the major risk factors of age and BMI were well matched between the arms of our study.
All this obfuscation clouds the very simple conclusion of the TARGIT-A trial. A regimen of risk adapted intraoperative radiotherapy analysed by intention to treat principles avoids the inconvenience and toxicity of whole breast radiotherapy in 80% of cases without any difference in cancer related outcomes. It also reduces the non-breast cancer mortality, and improves quality of life.
Finally, they even got the journal wrong – our paper was published in the BMJ, rather than BMJ Open.
References
1. Murray Brunt, A., et al., Hypofractionated breast radiotherapy for 1 week versus 3 weeks (FAST-Forward): 5-year efficacy and late normal tissue effects results from a multicentre, non-inferiority, randomised, phase 3 trial. Lancet, 2020. 395(10237): p. 1613-1626.
2. Vaidya, J.S., et al., Reduced Mortality With Partial-Breast Irradiation for Early Breast Cancer: A Meta-Analysis of Randomized Trials. Int J Radiat Oncol Biol Phys, 2016. 96(2): p. 259-265.
3. Vaidya, J.S., et al., Targeted radiotherapy for early breast cancer. Lancet, 2018. 391(10115): p. 26-27.
4. Sperk, E., et al., Late radiation toxicity after intraoperative radiotherapy (IORT) for breast cancer: results from the randomized phase III trial TARGIT A. Breast Cancer Res Treat, 2012. 135(1): p. 253-60.
5. Corica, T., et al., Cosmesis and Breast-Related Quality of Life Outcomes After Intraoperative Radiation Therapy for Early Breast Cancer: A Substudy of the TARGIT-A Trial. Int J Radiat Oncol Biol Phys, 2016. 96(1): p. 55-64.
6. Welzel, G., et al., Health-related quality of life after breast-conserving surgery and intraoperative radiotherapy for breast cancer using low-kilovoltage X-rays. Annals of surgical oncology, 2010. 17 Suppl 3: p. 359-67.
7. Coombs, N.J., et al., Environmental and social benefits of the targeted intraoperative radiotherapy for breast cancer: data from UK TARGIT-A trial centres and two UK NHS hospitals offering TARGIT IORT. BMJ Open, 2016. 6(5): p. e010703.
8. Pez, M., et al., Long-term outcome after intraoperative radiotherapy as a boost in breast cancer. Strahlenther Onkol, 2020. 196(4): p. 349-355.
9. Hughes, K.S., et al., Lumpectomy plus tamoxifen with or without irradiation in women 70 years of age or older with early breast cancer. N Engl J Med, 2004. 351(10): p. 971-7.
10. Kunkler, I.H., et al., Breast-conserving surgery with or without irradiation in women aged 65 years or older with early breast cancer (PRIME II): a randomised controlled trial. Lancet Oncol, 2015. 16(3): p. 266-73.
11. Blamey, R.W., et al., Radiotherapy or tamoxifen after conserving surgery for breast cancers of excellent prognosis: British Association of Surgical Oncology (BASO) II trial. Eur J Cancer, 2013. 49(10): p. 2294-302.
Competing interests: Same as the main paper
Dear Editor,
Firstly, we are very pleased to see the rapid response from one of the first independent patient advocates in the UK, Hazel Thornton who welcomes and applauds the publication of the long-term results of the TARGIT-A trial. She emphasises how “Women with early breast cancer can now make a better-informed decision on the basis of these updated findings, confident that intraoperative radiotherapy is an effective alternative to external beam radiotherapy.” We are very grateful for her input in the trial from its earliest days. We cannot overemphasise the great importance of the independent data monitoring committee and we sincerely apologise for the oversight of not acknowledging her contribution. Mrs Thornton has waged a decades-long campaign to make clinical evidence easily understood by patients. She recommends the use of pictograms such as the one we display in our figure 5, which gives clear information from the patient’s perspective.
Next we will respond here to the main issues raised by S Bentzen, et al, P Watson, S Maharjan et al, respectively.
To clear-up the misconception of Bentzen et al, the population enrolled in the TARGIT-A trial emphatically cannot be described as “low-risk”. In fact, the proportion of patients in various risk-groups is similar to that seen in our breast clinics, as is clear from Table 1 in the paper. For example, over 85% of patients were younger than 70, and the proportion of patients in “high risk” groups, as defined by tumour size, grade and lymph node status, were greater than the total populations of some of the early trials of partial breast irradiation (e.g. 366 were >2cm in size, 443 were grade 3 and 488 were node positive). So, by their lights, if we follow the statement of Bentzen et al that patients who received TARGIT-IORT alone did not need any radiotherapy to its logical conclusion, then 80% of patients in routine breast clinics should not receive any radiotherapy at all!
Next, the simple modelling by Bentzen et al rests on the assumption that almost all of the high-risk patients randomised to TARGIT-IORT also received supplemental EBRT. This is simply not true. The details will be fully revealed in an upcoming paper, but briefly and for example, supplemental EBRT was not given to 78% of Grade 3, not given to 82% of ER negative and not given 67% of node positive patients. So those who were treated by TARGIT-IORT alone (i.e. without any supplemental EBRT) were by no means all low risk. This important clinical detail invalidates their conclusion. Furthermore, by focussing only on local recurrence, their model also completely ignores the differential effect of the randomised treatments seen on mortality.
We have always recognised the theoretical points related to medical physics described by Dr Watson. As explained in our earlier response, the ultimate test of the technique is the clinical outcome – which clearly shows that TARGIT-IORT is an effective alternative to whole breast radiotherapy, and even reduces non-breast cancer mortality. Those are the robust data after long-term follow up – now, having access to these results, the onus for those working in the field of radiation biology and medical physics, would be to explain these counter-intuitive results. There has already been a lot of work in this field (cited in the paper) including the effects on tumour micro-environment, which could explain the success of the TARGIT-IORT approach.
With respect, we would like to dispel the misunderstanding of Maharjan and colleagues, that the group receiving supplemental whole breast radiotherapy were left out of the intention to treat (ITT) analysis. There was no selection bias whatsoever. It was efficacy of the policy of risk-adapted approach using TARGIT-IORT that was tested in the experimental arm. The criteria for recommending EBRT were specified in the protocol and treatment policy documents, and as explained above, only a small proportion of high-risk patients received supplemental EBRT after TARGIT-IORT. Secondly, we can clarify that higher risk group patients were not specifically recruited in later part of the trial. In fact, there were more node positive and ER negative cases in the earlier part of the trial.
With regard to the argument about improved radiotherapy techniques that have allegedly become safer in recent years – this point is only theoretical, never yet proven in any randomised trial.
As regards toxicity, we would like to reassure them that in fact there was lower radiation-related toxicity with TARGIT-IORT, and there was a slight 1.3% increase in the need for needle evacuation of post-operative seroma[1]. No higher incidence of fat necrosis was observed, which is hardly surprising since the TARGIT-IORT technique is gentle on tissues and does not require extensive dissection.
Reference
1. Vaidya, J.S., et al., Targeted intraoperative radiotherapy versus whole breast radiotherapy for breast cancer (TARGIT-A trial): an international, prospective, randomised, non-inferiority phase 3 trial. The Lancet, 2010. 376(9735): p. 91-102.
Competing interests: Same as the main paper
Re: Long term survival and local control outcomes from single dose targeted intraoperative radiotherapy during lumpectomy (TARGIT-IORT) for early breast cancer: TARGIT-A randomised clinical trial
Dear Editor,
John Yarnold et al`s rapid response [1], commenting on and illuminating the governance of the Targit-A trial report [2], has prompted me to belatedly respond to Jayant Vaidya et al`s rapid response of 18th September 2020 about this report. Vaidya`s reply was in response to my request for enlightenment in pursuit of clarity about how independent scrutiny had been achieved in the Targit-A trial follow up.
(https://www.bmj.com/content/370/bmj.m2836/rr-6)
Although it had been reassuring to learn from Vaidya and colleagues` response https://www.bmj.com/content/370/bmj.m2836/rr-11 [2] that the chairmanship of its Independent Steering Committee is in the reliable hands of Professor Freddie Hamdy, it did little to clarify how independent scrutiny had been achieved; rather the reverse, leaving me confused. If patients, citizens and doctors are to make well-informed decisions, we need to be confident that reports of findings are the best we can make them. [3]
Yarnold`s response [1] illuminates and begins to clarify the proper structures of committees required for conducting and overseeing research. His well-informed comment concerning the roles and responsibilities of governance structures, took my own attempt to understand which-committee-was-which a step further, and highlighted the omissions that occurred in the reports of the trials of Targit-A in the BMJ and elsewhere. It is comforting to realise that I was not alone in finding the reporting of the governance structures difficult to understand, incomplete and inadequate.
Surely, all trial reports should contain full details of DMCs and other governing committees and their Chairs? It is one thing to be confident in the data; yet another to be confident about accurate reporting of all those involved in the running and oversight of trials.
Perhaps this calls for the production of a checklist of titles/definitions/descriptions of committees involved in the set-up, running and reporting of trials so that authors and readers of trial reports and the ultimate beneficiary – patients/citizens – can be confident that the whole governance of a trial is unambiguous and transparent? The final stages of writing the report, submission and reviewing, and publication could advantageously work to this checklist to ensure that this was achieved consistently.
Mrs. Hazel Thornton, Hon.DSc. (Leicester)
Member of the Targit-A trial Data Monitoring Committee 12th July 2006 – 24th May 2013.
Independent Citizen Advocate for Quality in Research and Healthcare.
[1] Yarnold JR, Kirby A, Bentzen SM. BMJ rapid response to [2]
[2] Jayant S. Vaidya, Max Bulsara, Michael Baum, Frederik Wenz, Samuele Massarut, et al. Long term survival and local control outcomes from single dose targeted intraoperative radiotherapy during lumpectomy (TARGIT-IORT) for early breast cancer: TARGIT-A randomised clinical trial. BMJ 2020;370:m2836 https://www.bmj.com/content/370/bmj.m2836
[3] Thornton H. Better reporting of better research = better healthcare. EQUATOR NETWORK 3RD Annual Lecture. October 2011. https://www.equator-network.org/2011/10/03/better-reporting-of-better-re...
Competing interests: No competing interests