How will the CHHiP trial affect the future of prostate radiotherapy?

KEYWORDS:

• Prostate cancer
• radiotherapy
• hypofractionation
• clinical trials
• radiobiology
• changing practice

1. Introduction

P rostate cancer is the most common cancer in men in the United Kingdom, with 47, 151 new cases in 2015[1]. Management options include external-beam radiotherapy (EBRT), brachytherapy, radical prostatectomy, active surveillance (for men with low-risk disease), and watchful waiting (for those unsuitable for radical curative treatment), with management choices often affected by potential treatment-related side effects. EBRT is most appropriate for men with intermediate-risk or high-risk disease [2], and is associated with long-term disease control in most patients. High-dose conformal radiotherapy (CFRT) with conventional 2-Gy daily fractions to a total dose of 74 Gy has been the standard of care in the United Kingdom [3]. Meta-analysis has shown that high-dose radiotherapy (74–80 Gy) is associated with an increased risk of late gastrointestinal (GI) toxicity compared with lower doses of radiotherapy (64–70 · 2 Gy) [4]. Therefore, it is important to use advanced radiotherapy techniques such as intensity modulated radiotherapy (IMRT) or volumetric modified arc therapy (VMAT), which are able to sculpt dose distributions to the prostate target and image guided radiotherapy (IGRT) to make treatment as accurate as possible to avoid dose to surrounding normal tissues. As prostate cancer may be particularly sensitive to radiation fraction size [5], hypofractionated radiotherapy (HFRT), using fewer treatments of higher dose per fraction schedules, might improve the therapeutic ratio. HFRT could make treatment more effective and be safe and convenient for patients as well as using radiotherapy resource more efficiently. The impacts of clinical trials of HFRT are now being seen.

1.1. Clinical trials

Large phase 3 randomized controlled trials (RCTs) testing modest HFRT for localized prostate cancer have recently reported efficacy and side effect outcomes [6–9]. The two largest trials, CHHiP (3216 patients) [6,7] and PROFIT (1206 patients) [8], have much in common. Both studies compared standard fractionation (SFRT) schedules using 2.0-Gy daily fractions (total doses CHHiP 74 Gy; PROFIT 78 Gy) with HFRT schedules using 3.0-Gy daily fractions (total doses CHHiP 60 Gy and 57 Gy; PROFIT 60 Gy). Both trials tested the hypothesis that modest HFRT is non-inferior to SFRT in terms of disease control. IMRT with or without IGRT techniques was used in the both trials. The CHHiP trial showed that efficacy defined by biochemical/clinical failure-free outcome was non-inferior for 60 Gy compared with 74 Gy. At 5 years, the failure-free proportion was higher in the 60-Gy group at 90.6% (95% confidence intervals 88.5–92.3) compared with the 74-Gy group 88.3% (86.0–90.2) and 57-Gy group 85.9% (83.4–88.0). It was estimated that the 60 Gy in 3 Gy/fraction group had an approximate equivalence to 76 Gy/2 Gy fractions. The PROFIT trial included patients with intermediate risk disease and showed that 60 Gy was non-inferior to 78 Gy with identical 21% biochemical/clinical failure at 5 years. The 11% higher biochemical control rate in CHHiP is probably due to the routine use of androgen deprivation therapy (ADT); similar findings, a 13% gain in 5-year PSA control with 6 months of ADT added to radiotherapy, were reported in EORTC22991 [10]. Regarding side effects, the two trials also gave complimentary results. Both showed acute bowel (GI) and bladder (GU) symptoms peaked sooner with HFRT schedules (4–5 weeks) than SFRT (7–8 weeks) and there was a higher proportion of grade 2+ peak GI toxicity in both HFRT groups compared with SFRT. These differences disappeared by 18 weeks. In CHHiP there were no differences in long-term side effects between SFRT and HFRT groups at 5 years. Patient reported outcomes suggest an overall low incidence of GI and GU symptoms in all treatment groups; GI side effects were about 50% lower than in the previous MRC RT01 trial [6,7,11]. In PROFIT late GI side effects were increased in the 78 Gy SFRT group compared with HFRT group probably because of the relatively higher SFRT dose compared with CHHiP.

Additionally, the HYPRO study [9] evaluated the impact of a dose escalated HFRT schedule (see 1.5) and the RTOG 0415 trial tested HFRT in a low-risk population [12]. Impacts of these studies are now being seen in National and International treatment guidelines, change in radiotherapy practice, new clinical trial design and radiobiology research.

1.2. Guidelines

New treatment guidelines are being developed drawing on the results of these RCTs. Based on abstract presentations of the CHHiP Trial data the UK Royal College of Radiologists [13] recommended in December 2016 that acceptable regimens include 74–78 Gy in 37–39 fractions or 60 Gy in 20 fractions delivered using IMRT or VMAT techniques with IGRT verification. Following appraisal of full trial results NHS England have recently issued a Commissioning Policy [14] and concluded that HFRT is “safe and effective when delivered at 60 Gy/20 fraction schedule over a 4 week period” when compared to SFRT. Additionally, the 57 Gy/19 fraction schedule could be considered for frail patients who might tolerate side effects less well. A recent report of the over 75-year old sub-group in the CHHiP trial suggested the lower dose was better tolerated but also associated with good efficacy in this population [15]. NHS England will audit RT practice and expect 70% of patients treated with EBRT to receive HFRT. The German Society of Radiation Oncology commissioned an expert review of hypofractionation [16] which concluded that hypofractionation was appropriate provided strict quality control using IMRT/IGRT was employed as in U.K. guidance. These recommendations are endorsed by the European Association of Urology Guidelines 2017 Edition [17]. Presently, in North America, The National Comprehensive Cancer Network (NCCN) considers moderately hypofractionated (2.4–4.0 Gy schedules) using IMRT to be “an alternative to conventionally fractionated regimens when clinically indicated” [2]. The influential AUA/ASTRO/SUO group recommend that clinicians should consider moderate HFRT for localized prostate cancer patient (of any risk category) with a full appraisal of evidence and new guidance due in 2018 [18].

1.3. Changing clinical practice

The uptake of HFRT may well be influenced by different health care systems, re-imbursement methods and patient and clinician perspectives. In the United Kingdom, the National Cancer Registration and Analysis Service (NCRAS) has recorded prostate cancer fractionation robustly in the National Radiotherapy Dataset (RTDS) [19]. There has been an increase in modest hypofractionation (treatment in 20 fractions) from 8%, 15%, 17%, 21% rising to 49% in 2012–2013, 2013–2014, 2015–2016, and 2016–2017, respectively, with a corresponding fall of high dose SFRT (74–78 Gy in 37–39 fractions) from 77%, 62%, 60%, 45%, and 26%. The changes temporally correspond to the preliminary and 5-year reports of the CHHiP trial in 2012 and 2016, respectively [6,20].

Impact on resource use can be estimated at about £28 M/annum assuming: (1) the 70% target is met in the United Kingdom (from a baseline use of 8%), (2) 17 fewer fractions of radiotherapy are given to 62% of the 16,000 patients treated radically each year [19] with (3) a tariff charge of £177 for each fraction of IMRT/IGRT. Patients will derive additional benefits from fewer appointments and reduced travel costs.

1.4. Clinical trial design

In the United Kingdom, new trials have adopted modest HFRT as the new “standard of care” and the 60 Gy in 20 fraction schedule is used in the PIVOTALboost phase 3 trial (ISRCTN80146950) testing the addition of (a) pelvic lymph node radiotherapy and (b) focal boosts to intra-prostatic tumor nodules. The PACE trial (ISRCTN17627211) which tests an extreme hypofractionation schedule (36.25 Gy in five fractions) has been modified so that modest HFRT rather than SFRT is the standard comparator group. The phase 2 DELINEATE trial (ISRCTN04483921) studying the safety of intra-prostatic nodule boosts has tested an HFRT cohort.

1.5. Radiobiology

The fraction sensitivity of cancer and normal tissues is usually described by a linear quadratic equation. The ratio (α/β) of the linear (α) and quadratic (β) terms is a useful description of the shape of dose-effect curves. Early responding tissues and many cancers typically have a high α/β ratio, making them sensitive to total dose but with a small fractionation effect. Late responding tissues and perhaps prostate cancer have a low α/β ratio, demonstrating increased survival at low dose/fraction and significantly greater toxicity at higher dose/fraction.

Simplistically, the CHHiP and PROFIT Trials suggest a low α/β ratio of 1.8–1.4 Gy. However a third phase 3 trial has suggested that an additional time factor may be important [13]. The HYPRO trial [9], randomized 804 patients and compared SFRT using 78 Gy in 2-Gy daily fractions with HFRT giving 64 Gy in 3 · 4 Gy fractions but importantly treating with three fractions/week prolonging the HFRT by about 2 weeks compared to CHHiP and PROFIT. The gain in tumor control was smaller than might have been expected (HFRT 80.5% vs. CFRT 77.1% at 5 years) and not statistically significant. This could be the play of chance but a biological explanation may be the relative treatment protraction and if a time factor is included the best estimate α/β ratio might increase to 4–5 Gy [21] making it similar or higher than critical normal tissues. This is potentially important for the introduction of more extreme forms of hypofractionation. The recently completed HYPO (testing 42.7 Gy in seven fractions; ISRCTN45905321) and PACE-B phase 3 trials will provide invaluable data in coming years.

The results of RCTs will continue to guide our understanding of how best to deliver treatment and understand the radiobiology of prostate cancer. The hope for the future is that very safe, highly effective and rapidly completed radiotherapy will become available for patients with prostate cancer.

Declaration of interest

D Dearnaley reports receiving personal fees and acting on the advisory board and consulting for Takeda, Amgen, Janssen, Astellas and Sandoz. He also reports receiving grants from Cancer Research UK and National Institute for Health Research. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed. Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This manuscript has received funding from Cancer Research UK (A21518, C46/A10588, C33589/A19727, C8262/A7253, C1491/A9895, C1491/A15955, SP2312/021) and the National Institute of Health Research Royal Marsden Hospital and the Institute of Cancer Research Biomedical Research Centre.