Clinical Oncology (2014) e21ee46 Contents lists available at ScienceDirect

Clinical Oncology journal homepage: www.clinicaloncologyonline.net

Original Article

Hormone and Radiotherapy versus Hormone or Radiotherapy Alone for Non-metastatic Prostate Cancer: A Systematic Review with Meta-analyses M. Schmidt-Hansen *, P. Hoskin y, P. Kirkbride z, E. Hasler *, N. Bromham * * National

Collaborating Centre for Cancer, Cardiff, UK Mount Vernon Cancer Centre, Northwood, Middlesex, UK z The Clatterbridge Cancer Centre NHS Foundation Trust, Bebington, Wirral, UK y

Received 6 January 2014; received in revised form 28 May 2014; accepted 3 June 2014

Abstract Aims: Radiotherapy is standard treatment for localised prostate cancer and is often combined with hormone treatment to prevent androgen stimulation of prostate cancer. Hormone therapy carries significant morbidity and can only be justified in the radical treatment of localised disease if it can be balanced against a significant gain in disease control and survival. Materials and methods: We searched Medline, Premedline, Embase, Cochrane Library, Web of Science (SCI & SSCI) and Biomed Central for randomised controlled trials published in English comparing radiotherapy or hormone therapy alone with radiotherapy and hormone therapy in combination as first-line treatment in patients with non-metastatic prostate cancer reporting overall survival, disease-free survival, distant metastases-free survival, biochemical survival, adverse events (including cardiovascular) and/or health-related quality of life. Results: Fourteen trials were included and showed that combination therapy was associated with better or similar survival and disease-free outcomes compared with single-modality treatment, and that this may particularly be the case for patients with higher risk disease. The results also suggested that combination therapy is associated with more and worse adverse events and quality of life, although this was not always the case. Some of the results are at risk of reporting bias. Conclusion: The published data support the use of combined treatment with androgen deprivation and radiotherapy for intermediate- and high-risk localised and locally advanced prostate cancer. Optimal timing, duration, formulation and the management of side-effects remain important questions for further research. Ó 2014 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

Key words: First-line treatment; hormone therapy; locally advanced; prostate cancer; prostate neoplasm; radiotherapy

Introduction High-dose external beam radiotherapy delivered using modern intensity-modulated and image-guided techniques is a standard treatment for localised prostate cancer. It is usually combined with a period of androgen deprivation therapy (ADT), using drugs that prevent androgen stimulation of prostate cancer by blocking the luteinising hormone releasing hormone (LHRH) release controlling the hypothalamicepituitary axis release of follicle-stimulation hormone (FSH) and luteinising hormone (LH), which Author for correspondence: M. Schmidt-Hansen, National Collaborating Centre for Cancer, Park House, Greyfriars Road, Cardiff CF10 3AF, UK. Tel: þ44-2920-402916; Fax: þ44-2920-402911. E-mail address: [email protected] (M. Schmidt-Hansen).

stimulate the testes to produce testosterone. This simulates the effect of surgical castration. Alternatively anti-androgen drugs acting as competitive antagonists of androgen receptor activity may be used. In the metastatic setting, combinations of these two approaches achieve maximal androgen blockade. Although most trials of hormone therapy used in association with radiotherapy have used ADT, some have used anti-androgens. Hormone therapy carries significant morbidity and can only be justified in the radical treatment of localised disease if it can be balanced against a significant gain in disease control and survival. The mechanism of androgen suppression synergy with radical local radiotherapy is uncertain, but may include both sensitisation of the cancer cell to radiation and modification of the metastatic process [1]. Initial response in terms of

http://dx.doi.org/10.1016/j.clon.2014.06.016 0936-6555/Ó 2014 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

e22

M. Schmidt-Hansen et al. / Clinical Oncology (2014) e21ee46

prostate-specific antigen (PSA) reduction is almost universal, apart from very poorly differentiated tumours, which may not be considered for radical local treatment. Thus, a reduction in the number of viable tumour cells at the primary site might be expected, and so ablative local therapy would have a higher chance of eradicating sufficient cell numbers to achieve cure. In addition, hormone therapy is a systemic treatment and could affect micrometastases, which may be more important when seeking a survival effect. It has been suggested that within a short time of starting ADT there is significant vascular shutdown within the tumour [2], which theoretically could invoke radioresistance through hypoxia, but this has never been shown to affect clinical outcome of combined treatment. Radiotherapy is a defined event within a specific timeframe, typically 7e8 weeks. Hormone therapy may be given for a variable length of time and may precede radiotherapy (neoadjuvant treatment, NAH), be given during radiotherapy and for a period after radiotherapy. The optimal timing and overall duration is uncertain; typically, patients with ‘intermediate- to high-risk’ localised disease receive NAH for 3e6 months before radiotherapy, whereas patients with ‘high-risk’ or locally advanced cancers might receive hormone treatment for 2 years or longer, with NAH often, but not always, part of that treatment. Which patients should receive hormone therapy, when, what type and for how long have not been clearly defined. The aim of this study was to compare the outcomes of patients who have received external beam radiotherapy and hormone therapy, singly or in combination, as first-line treatment for prostate cancer, and to examine whether certain patient risk groups benefit from any of the treatment strategies.

Materials and Methods Criteria for Considering Studies in this Review Randomised controlled trials published in English comparing radiotherapy or hormone therapy alone with radiotherapy and hormone therapy in combination as firstline treatment in patients with non-metastatic prostate cancer reporting overall survival, disease-free survival, distant metastases-free survival, biochemical survival, adverse events (including cardiovascular) and/or healthrelated quality of life were considered. Search Methods for Identification of Studies The search strategy consisted of two searches; one updating the Cochrane review by Kumar et al. on this topic [3]; thus conducted from 2006 to 16 May 2013 for Medline, Premedline, Embase, Cochrane Library, Web of Science SCI & SCII and Biomed Central and an additional full search for orchiectomy (Medline [1946 to 16 May 2013], Premedline [15 May 2013], Embase [1974 to 16 May 2013], Cochrane Library [16 May 2013], Web of Science [SCI 1899 to 16 May 2013 and SSCI; 1956 to 16 May 2013] and Biomed Central [as

per database]) using the OVID Medline search strategy (adapted to each database) detailed in Table 1. Data Analysis One author screened the results of the computerised search and imported all potentially relevant papers into a database, which was screened for relevant studies by another author. The inclusion of potentially relevant studies was confirmed by consensus between three of the authors Table 1 OVID Medline search strategy (adapted to each database) 1 exp Prostatic Neoplasms/ 2 Prostatic Intraepithelial Neoplasia/ 3 (prostat$ adj3 (cancer$ or carcinoma$ or adeno$ or malignan$ or tum?r$ or neoplas$ or intraepithelial$)).tw. 4 PIN.tw. 5 1 or 2 or 3 or 4 6 exp Radiotherapy/ 7 Radiotherapy, Adjuvant/ 8 radiotherap$.tw. 9 (radiation adj (therap$ or treatment$)).tw. 10 external beam irradiation.tw. 11 external beam therap*.tw. 12 external beam treatment*.tw. 13 (EBRT or XRT).tw. 14 (CRT or 3DCRT or IMRT).tw. 15 conformal irradiation.tw. 16 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 17 5 and 16 18 exp Antineoplastic Agents, Hormonal/ 19 exp Androgen Antagonists/ 20 antiandrogens.mp. 21 ((androgen$ or hormon$) adj3 (ablat$ or block$ or withdraw$ or depriv$ or supress$)).mp. 22 gonadotrophin releasing hormone analogue$.mp. 23 (luteinizing hormone releasing hormone or LHRH).mp. 24 grha.tw. 25 (zoladex or decapeptide).mp. 26 (eligard or leuprorelin or enatone or a-43818 or lupron or tap144).mp. 27 exp Gonadotropin-Releasing Hormone/ 28 exp Cyproterone/ 29 (bicalutamide or casodex).mp. 30 exp Estrogens/ 31 oestrogen.mp. 32 exp Flutamide/ 33 (niftolid$ or eulexin).mp. 34 (nilutamide or nilandron$).mp. 35 exp Diethylstilbestrol/ 36 exp Progestins/ 37 exp Finasteride/ 38 proscar.mp. 39 adjuvant hormon$ therap$.tw. 40 (neoadjuvant or neo-adjuvant hormon$ therap$).tw. 41 17 and 40 42 limit to 2006 onwards 43 exp Orchiectomy/ 44 (orchiectom$ or orchidectom$).tw. 45 43 or 44 46 17 and 45 47 42 or 46

M. Schmidt-Hansen et al. / Clinical Oncology (2014) e21ee46

and one of the authors extracted the data and assessed the quality of the studies using the Cochrane Collaboration quality checklist for randomised controlled trials [4]. The time-to-event data were meta-analysed in Review Manager 5.2 [5] using the Exp[(O-E)/Var] statistical method and a fixed effect model after extracting the data according to the methods described by Tierney et al. [6]. The cardiovascular and adverse events and health-related quality of life data were summarised narratively. Subgroup analyses were carried out by treatment and, where possible, risk group.

Results Search Results The search of all the databases identified 2953 (before deduplication) possibly relevant papers, of which 2873 papers were excluded based on title/abstract; 80 papers were obtained for full-text review. Fifty of these 80 papers were excluded as they did not meet the inclusion criteria and 14 trials published in 30 papers were included in this review. Characteristics and Findings of the Studies Table 2 provides a summary of the treatment groups of each trial and of the cardiovascular and adverse events as well as health-related quality of life observed in the trials. Table 3 summarises the risk-of-bias assessments for each of the included trials and Appendix 1 provides a detailed description and assessment of each trial. Overall Survival Radiotherapy Alone versus Radiotherapy Plus Hormone Therapy Figure 1 illustrates the results from the individual studies grouped by treatment and their meta-analysis and shows that overall survival was better in the patients who received combination therapy compared with those receiving only radiotherapy when hormone therapy was adjuvant (hazard ratio ¼ 1.32; 95% confidence interval 1.17e1.47; P < 0.00001; I2 ¼ 41%), neoadjuvant (hazard ratio ¼ 1.25; 95% confidence interval 1.12e1.39; P < 0.0001; I2 ¼ 39%), neoadjuvant, concomitant and adjuvant (hazard ratio ¼ 1.72; 95% confidence interval 1.25e2.39; P ¼ 0.001; I2 ¼ 0%; although Fellows et al. [36] reported that at 7 years, 67/88 radiotherapyalone patients and 64/99 combination therapy patients had died, a difference that was not statistically significant) and when all the studies were combined regardless of the order of combination treatment (hazard ratio ¼ 1.3; 95% confidence interval 1.2e1.41; P < 0.00001; I2 ¼ 36%). Hormone Therapy Alone versus Radiotherapy During Hormone Therapy Of the four trials, only PR07 [30e33] reported the results such that they could be entered into a meta-analysis. Fellows et al. [36], Mottet et al. [28,29] and SPCG-7/SFUO-3 [34,35] found that overall survival did not differ significantly

e23

between the groups at 7 (Fellows et al. [36], SPCG-7/SFUO-3 [34,35]) or 5.6 years (Mottet et al. [28,29]) reporting 35, 71.4 and 83.5%, respectively, for the combination therapy groups and 36, 71.5 and 79.9%, respectively, for the hormone therapy-alone groups. However, at a median follow-up time of 8 years, PR07 [30e33] reported that overall survival was superior in the combination therapy group, relative to the hormone therapy-alone group (hazard ratio ¼ 1.43; 95% confidence interval 1.17e1.74; P ¼ 0.0005), and similarly at 10 years, SPCG-7/SFUO-3 [34,35] found superior survival in the combination therapy group relative to the singlemodality treatment group (absolute risk difference ¼ 9.8%; 95% confidence interval 0.8e18.8%; relative risk of overall death ¼ 0.68; 95% confidence interval 0.52e0.89; P ¼ 0.004). Disease-free Survival Radiotherapy Alone versus Radiotherapy and Adjuvant Hormone Therapy Figure 2 illustrates the results from the individual studies, which all favoured combination therapy statistically significantly (EORTC 22863 [7,8], EPC23-25 [9,10] locally advanced, RTOG 85-31 [11e13]) or numerically, but non-statistically significantly (EPC23-25 [9,10] localised; Zagars et al. [14]). Although the meta-analysis also showed that combination therapy was associated with longer disease-free survival, the between-study heterogeneity was so high (I2 ¼ 81%) that this analysis should be interpreted with extreme caution. Radiotherapy Alone versus Neoadjuvant Hormone Therapy and Radiotherapy Figure 2 shows the results of the two (RTOG 86-10 [15,16], TROG 96.01 [17,18] 6 month hormone treatment group) of the four studies that reported this outcome and their metaanalysis, which showed that disease-free survival was longer in the patients who received combination therapy compared with those receiving only radiotherapy (hazard ratio ¼ 1.47; 95% confidence interval 1.28e1.68; P < 0.0001; I2 ¼ 0%). Radiotherapy Alone versus Neoadjuvant, Concomitant and Adjuvant Hormone Therapy Plus Radiotherapy Only one of the four trials reported this outcome. Granfors et al. [21,22] found that disease-free survival was longer with combination therapy than with radiotherapy-alone treatment (hazard ratio ¼ 2.51; 95% confidence interval 1.32e4.76; P ¼ 0.005; see Figure 2). Radiotherapy Alone versus Radiotherapy Plus Hormone Therapy (in any order) When the results of all the studies that compared radiotherapy alone with combination therapy were metaanalysed, the between-study heterogeneity was high (71%). The individual study results were therefore not combined in an overall analysis. Hormone Therapy Alone versus Radiotherapy During Hormone Therapy Of the four trials, Fellows et al. [36] and SPCG-7/SFUO-3 [34,35] did not report this outcome and Mottet et al.

e24

Table 2 Summary of the treatment groups of each trial and of the cardiovascular and adverse events as well as health-related quality of life observed in the trials Trial [reference]

Treatment groups

Radiotherapy alone versus radiotherapy and adjuvant hormone therapy EORTC RT: n ¼ 208, 70 Gy; 22863 [7,8] HT/RT: n ¼ 207, 70 Gy þ goserelin acetate and cyproterone acetate (3 years) EPC23-25 RT: n ¼ 607, w64 Gy; HT/RT: [9,10] n ¼ 699, w64 Gy þ bicculetamide (2 years)

TROG 96.01 [17,18]

RT: n ¼ 270, 66 Gy; HT3/RT: n ¼ 265, 66 Gy þ goserelin acetate and flutamide (3 months); HT6/RT:

Adverse events

Health-related quality of life

Cardiovascular mortality: no difference

Not reported

Not reported

Cardiovascular mortality: no difference

Diarrhoea: HT/RT ¼ 15.6%, RT ¼ 14%; asthenia: HT/RT ¼ 13.5%, RT ¼ 9.8%; impotence: HT/RT ¼ 12.7%, RT ¼ 9.9%; decreased libido: HT/RT ¼ 4%, RT ¼ 1.4%; hot flashes: HT/RT ¼ 9.8%, RT ¼ 5.4%; back pain: HT/RT ¼ 12%, RT ¼ 13.9%; pharyngitis: HT/RT ¼ 11.4%, RT ¼ 11.1%; rectal haemorrhage: HT/RT ¼ 11.4%, RT ¼ 11.3%; constipation: HT/RT ¼ 11.1%, RT ¼ 9.2%; rash: HT/RT ¼ 10.8%, RT ¼ 8.9%; haematuria: HT/RT ¼ 9.5%, RT ¼ 12.7%; arthralgia: HT/RT ¼ 8.6%, RT ¼ 11.1%; abnormal liver function: HT/RT ¼ 2.2%, RT ¼ 1.8%; withdrawal due to adverse events: HT/RT ¼ 31.4%, RT ¼ 11%. In the HT/RT patients the most common adverse events were breast pain (75.1%) and gynecomastia (67.3%). These were mildemoderate in >90% cases. Not reported

Not reported

Not reported

Not reported

Not reported

Cardiac death: no difference found

Acute (grade 3): HT/RT ¼ 2%, RT ¼ 4%; late (grade 3): 8% in both groups; late (grade 4): HT/RT ¼ 1%, RT ¼ 3%; 53/264 HT3/RT patients and 71/267 HT6/RT patients

Not reported

Cardiovascular mortality: No difference

Cardiac death: no difference found

M. Schmidt-Hansen et al. / Clinical Oncology (2014) e21ee46

RT: n ¼ 468, 60e70 Gy; HT/RT: n ¼ 477, 60e70 Gy þ goserelin acetate (indefinitely/until disease progression) Zagars RT: n ¼ 43, 70 Gy; et al. [14] HT/RT: n ¼ 39, 70 Gy þ diethylstilbestrol (indefinitely) Radiotherapy alone versus neoadjuvant hormone therapy and radiotherapy RTOG RT: n ¼ 232, 65e70 Gy; HT/RT: n ¼ 224, 86-10 [15,16] 65e70 Gy þ goserelin acetate and flutamide (4 months)

RTOG 85-31 [11e13]

Cardiovascular events

Not reported

Sexual activity rates: Baseline:

n ¼ 267, 66 Gy þ goserelin acetate and flutamide (6 months)

L-101 [19]

Radiotherapy alone versus neoadjuvant, concomitant and adjuvant hormone therapy followed by radiotherapy L-101 [19] RT: n ¼ 43, 64 Gy; HT/RT: n ¼ 55, 64 Gy þ a LHRH agonist plus an anti-androgen agent (10 months) Granfors RT: n ¼ 46, 65.2 Gy; HT/RT: n ¼ 45, 64.9 Gy et al. [21,22] þ orchiectomy (indefinitely) D’Amico RT: n ¼ 104, 70.35 Gy; HT/RT: n ¼ 102, 70.35 Gy et al. [23e27] þ leuprolide acetate, goserelin acetate and flutamide (3 months)

RT ¼ 37.5%, HT3/RT ¼ 33.3%, HT6/RT ¼ 36%; Post-RT: RT ¼ 18.6%, HT3/RT ¼ 3%, HT6/RT ¼ 2.6%; 1 year post-RT: RT ¼ 18.6%, HT3/RT ¼ 12.1%, HT6/RT ¼ 14.2% Not reported

Not reported

Not reported

Acute hepatic (grade 3e4): HT/RT ¼ grade 3): diarrhoea: HT/RT: n ¼ 8, HT: n ¼ 4; rectal bleeding: HT/RT: n ¼ 2, HT: n ¼ 3; genitourinary: both groups: n ¼ 14; late proctitis (>grade 2): HT/RT ¼ 1%, HT ¼ 0.3%.

Not reported

PR07 [30e33]

HT: n ¼ 602, LHRH agonist or orchiectomy (indefinitely); HT/RT: n ¼ 603, 65e69 Gy þ LHRH agonist or orchiectomy (indefinitely)

Not reported

SPCG-7/SFUO-3 [34,35]

HT: n ¼ 439, leuprorelin and flutamide (3 months); HT/RT: n ¼ 436, 70 Gy þ leuprorelin and flutamide (3 months)

Not reported

5 years: bladder obstruction/sclerosis, urinary frequency per 24 hours (>10), intestinal symptoms (moderate/severe), sexual activity: HT/RT ¼ HT; urethral stricture, urgency, urinary incontinence, and erection problems: HT/RT > HT; serious adverse events: HT/RT: n ¼ 7, HT: n ¼ 11

6 months: overall quality of life and physical functioning: HT > HT/RT; diarrhoea, bowel or rectal and urinary functioning: HT < HT/RT; 36 months: all categories: HT ¼ HT/RT 4 years: physical /role/emotional /cognitive function, global health /quality of life, fatigue, nausea /vomiting, pain, dyspnoea, insomnia, appetite loss, constipation, and financial difficulties: HT ¼ HT/RT; Social function and diarrhoea: HT > HT/RT

M. Schmidt-Hansen et al. / Clinical Oncology (2014) e21ee46

Cardiovascular events

Fellows et al. [36] *

HT: n ¼ 90, orchiectomy (indefinitely); RT: n ¼ 88, Gy not specified; HT/RT: n ¼ 99, Gy not specified þ orchiectomy (indefinitely)

Not reported

Overall HT-related: hot flushes (n ¼ 28), rectal bleeding, frequency and urgency of micturition (all n ¼ 1); overall RT-related: bowel symptoms (n ¼ 35), urinary symptoms other than transient frequency (n ¼ 15), both (n ¼ 2), severe rectal bleeding (n ¼ 4). n ¼ 2 who died had radiation proctitis.

Not reported

RT, radiotherapy; HT, hormone therapy. * It is unclear in which order the hormone and radiotherapy treatments were given in the combined treatment group in this trial.

Selection bias

D’Amico et al. [23e27] EORTC 22863 [7,8] EPC23-25 [9,10] Fellows et al. [36] Granfors et al. [21,22] Jones et al. [20] L-101 [19] Mottet et al. [28,29] PR07 [30e33] RTOG 85-31 [11e13] RTOG 86-10 [15,16] SPCG-7/SFUO-3 [34,35] TROG 96.01 [17,18] Zagars et al. [14]

Detection bias

Attrition bias

Reporting bias

Random sequence generation

Allocation concealment

Blinding of outcome assessment: objective outcomes

Blinding of outcome assessment: subjective outcomes

Incomplete outcome data

Selective reporting

þ þ ? þ ? þ ? ? þ þ ? þ þ ?

þ þ ? þ ? ? ? þ þ ? ? þ þ ?

? ? ? ? ? ? ? e e ? ? e ? ?

? ? ? ? ? ? ? e e ? ? e ? ?

þ þ þ þ þ þ þ þ þ þ þ þ þ þ

? ? þ þ ? þ e þ þ ? þ ? ? ?

e27

þ, low risk of bias; ?, unclear risk of bias; e, high risk of bias.

M. Schmidt-Hansen et al. / Clinical Oncology (2014) e21ee46

Table 3 Risk of bias assessment for the included trials

e28

M. Schmidt-Hansen et al. / Clinical Oncology (2014) e21ee46

Fig 1. Overall survival. The events entered are the number of deaths. For EPC23-25 [9,10], the hazard ratios used to calculate OeE seem to be adjusted for trial, randomised treatment, initial prostate-specific antigen level, tumour grade and stage, whereas for the rest of the studies, unadjusted estimates were used. The data entered for TROG 96.01 [17,18] are for RT alone versus 6 months HT þ RT. TROG 96.01 [17,18] also compared RT alone with 3 months of HT þ RT (10 year all-cause mortality ¼ 42.5% for RT alone and 36.7% for HT3/RT; P ¼ 0.2). 0, not reported; RT, radiotherapy; HT, hormone therapy.

[28,29] found that 43/130 hormone therapy-alone patients and 13/133 combination therapy patients experienced clinical disease progression, but did not report these data such that they could be included in the meta-analysis. PR07 [30e33] found that combination therapy was associated with longer disease-free survival than hormone therapy alone (hazard ratio ¼ 3.33; 95% confidence interval 2.56e4.34; P < 0.00001). Distant Metastases-free Survival Radiotherapy Alone versus Radiotherapy Plus Hormone Therapy Figure 3 illustrates the results from the individual studies grouped by treatment and their meta-analysis and shows that distant metastases-free survival was better in the patients who received combination therapy compared with those receiving only radiotherapy when hormone therapy was adjuvant (hazard ratio ¼ 1.73; 95% confidence interval 1.46e2.06; P < 0.00001; I2 ¼ 47%), neoadjuvant (hazard ratio ¼ 1.49; 95% confidence interval 1.22e1.82; P ¼ 0.0001; I2 ¼ 0%) and overall (hazard ratio ¼ 1.63; 95% confidence

interval 1.43e1.85; P < 0.00001; I2 ¼ 0%). Of the four trials that examined radiotherapy alone versus neoadjuvant, concomitant and adjuvant hormone therapy þ radiotherapy, only Fellows et al. [36] reported distant disease-free survival, but not in a manner that allowed it to be included in the meta-analysis. Fellows et al. [36] found that combination therapy was associated with significantly longer distant metastases-free survival at 7 years, with 71/88 radiotherapy-alone patients and 59/99 combination therapy patients having distant metastases. Hormone Therapy Alone versus Radiotherapy During Hormone Therapy Fellows et al. [36] found no difference in distant metastasis-free survival at 7 years between the hormone therapy-alone group (61%) and the combination therapy group (60%), whereas Mottet et al. [28,29] reported that distant metastasis-free survival was longer in the combination therapy group than in the hormone therapy group (hazard ratio ¼ 3.82; 95% confidence interval 1.26e11.69; P ¼ 0.02). PR07 [30e33] and SPCG-7/SFUO-3 [34,35] did not report this outcome.

M. Schmidt-Hansen et al. / Clinical Oncology (2014) e21ee46

e29

Fig 2. Disease-free survival: The events entered are the number of failures. For EPC23-25 [9,10], the hazard ratios used to calculate OeE seem to be adjusted for trial, randomised treatment, initial prostate-specific antigen level, tumour grade and stage, whereas for the rest of the studies, unadjusted estimates were used. The data entered for TROG 96.01 [17,18] are for RT alone versus 6 months HT þ RT. TROG 96.01 [17,18] also compared RT alone with 3 months of HT þ RT (10 year disease-free survival ¼ 12.7% for RT alone and 28.8% for HT3/RT; P < 0.0001). Heterogeneity is high overall and in the radiotherapy alone versus radiotherapy and adjuvant hormone therapy subgroup; this may be due to combining data from different risk groups. 0, not reported; RT, radiotherapy; HT, hormone therapy.

Biochemical Disease-free Survival Radiotherapy Alone versus Radiotherapy and Adjuvant Hormone Therapy This outcome was only reported by one of the four trials (EPC23-25 [9,10]). Figure 4 illustrates the results from this study for patients with localised and locally advanced disease, which both favoured combination treatment statistically significantly. Although the metaanalysis also showed that combination therapy was associated with longer biochemical disease-free survival, the unexplained heterogeneity was so high (I2 ¼ 91%) that this analysis should be interpreted with extreme caution. Radiotherapy Alone versus Neoadjuvant Hormone Therapy and Radiotherapy Figure 4 shows the results of all the individual studies (data used from the 6 month hormone treatment group from TROG 96.01 [17,18]) and their meta-analysis, which showed that biochemical disease-free survival was longer in the patients who received combination therapy compared with those receiving only radiotherapy (hazard ratio ¼ 1.65; 95% confidence interval 1.48e1.83; P < 0.00001; I2 ¼ 0%).

Radiotherapy Alone versus Neoadjuvant, Concomitant and Adjuvant Hormone Therapy þ Radiotherapy Figure 4 shows the results of the two studies (D’Amico et al. [23e27], L-101 [19] data included for the group receiving 10 months of hormone therapy) that reported this outcome together with their meta-analysis, which all showed that also in the case of biochemical disease-free survival is combination treatment associated with a more favourable outcome than treatment with radiotherapy alone (hazard ratio ¼ 2.53; 95% confidence interval 1.75e3.67; P < 0.00001; I2 ¼ 0%). Radiotherapy Alone versus Radiotherapy Plus Hormone Therapy (in any order) Although the consistent finding that combination therapy is associated with longer biochemical disease-free survival than radiotherapy treatment alone was confirmed when the results of all the studies that compared radiotherapy alone with combination therapy were metaanalysed, this analysis should be interpreted with caution because there was substantial between-study heterogeneity and the data from the radiotherapy-alone group in L-101 [19] is entered twice as a comparator for the two combination treatment groups (hazard ratio ¼ 1.67; 95% confidence interval 1.54e1.82; P < 0.00001; I2 ¼ 54%).

e30

M. Schmidt-Hansen et al. / Clinical Oncology (2014) e21ee46

Fig 3. Distant metastasis-free survival. The events entered are the number of failures. The data entered for TROG 96.01 [17,18] are for RT alone versus 6 months HT þ RT. TROG 96.01 [17,18] also compared RT alone with 3 months of HT þ RT (10 year distant progression ¼ 13.5% for RT alone and 14.5% for HT3/RT; P ¼ 0.82). 0, not reported; RT, radiotherapy; HT, hormone therapy.

Hormone Therapy Alone versus Radiotherapy During Hormone Therapy Fellows et al. [36] did not report this outcome and the remaining three trials did not report this outcome in a manner that allowed the data to be meta-analysed. Mottet et al. [28,29] found that the PSA progression rate was higher in the hormone therapy-only patients compared with the combination treatment patients (78.5% or 68.5% versus 17.3% or 14.3%, depending on which criteria were used). PR07 [30e33] reported that 20% of the hormone therapy patients and 7% of the combination therapy patients experienced biochemical relapse, and SPCG-7/SFUO-3 [34,35] found that combination therapy was associated with a significantly lower risk of PSA recurrence than hormone therapy alone, both at 7 (absolute risk reduction ¼ 53.5%; 95% confidence interval 47.3e59.7%) and 10 years (absolute risk reduction ¼ 48.8%; 95% confidence interval 40.4e57.2%; relative risk ¼ 0.16; 95% confidence interval 0.12e0.2; P < 0.0001). Additional Analyses Based on Study-reported Risk Groups Radiotherapy Alone versus Radiotherapy and Adjuvant Hormone Therapy EORTC 22863 [7,8] reported an overall survival advantage for the combination treatment group when the analysis was restricted to T3e4 patients (90% of the whole

sample; hazard ratio ¼ 0.56; 95% confidence interval 0.41e0.75, P ¼ 0.0001; 10 year overall survival rates ¼ 58.8% for radiotherapy/hormone therapy and 37.7% for radiotherapy). EPC23-25 [9,10] also reported such an advantage, but only for the patients with locally advanced disease (hazard ratio ¼ 1.54; 95% confidence interval 1.06e2.24), not those with localised disease (hazard ratio ¼ 1.06; 95% confidence interval 0.84e1.34) and, similarly, RTOG 85-31 [11e13] observed that combination treatment was associated with longer overall survival in patients with a Gleason score of 7 (hazard ratio ¼ 1.38; 95% confidence interval 1.04e1.82) or 8e10 (hazard ratio ¼ 1.49; 95% confidence interval 1.13e1.97), but not 2e4 (hazard ratio ¼ 1.22; 95% confidence interval 0.88e1.7). In terms of disease-free survival, EPC23-25 [9,10] also observed an advantage for combination therapy, but only in patients with locally advanced disease (hazard ratio ¼ 1.79; 95% confidence interval 1.3e2.46) and not localised disease (hazard ratio ¼ 1.16; 95% confidence interval 0.94e1.44), and RTOG 85-31 [11e13] also found an advantage in disease-free survival of combination treatment when the analysis was restricted to patients with PSA < 1.5 ng/ml ([radiotherapy/ hormone therapy: 126/440] > [radiotherapy: 46/431], P < 0.0001). Although EPC23-25 [9,10] reported that combination therapy was associated with longer biochemical disease-free survival in both patients with localised (hazard

M. Schmidt-Hansen et al. / Clinical Oncology (2014) e21ee46

e31

Fig 4. Biochemical-free survival. The events entered are the number of failures. For EPC23-25 [9,10], the hazard ratios used to calculate OeE seem to be adjusted for trial, randomised treatment, initial prostate-specific antigen level, tumour grade and stage, whereas for the rest of the studies, unadjusted estimates were used. The data entered for TROG 96.01 [17,18] are for RT alone versus 6 months HT þ RT. TROG 96.01 [17,18] also compared RT alone with 3 months of HT þ RT (10 year prostate-specific antigen progression ¼ 73.8% for RT alone and 60.4% for HT3/RT; P ¼ 0.0009). The data from D’Amico et al. [23e27] are based on n ¼ 103 and 98 instead of n ¼ 104 and 102, for the RT and RTR/HT groups, respectively. No overall estimate is provided because the same data from the RT-alone group in the study by L-101 [19] are used in two subgroups. Heterogeneity is high in the radiotherapy alone versus radiotherapy and adjuvant hormone therapy subgroup; this may be due to combining data from different risk groups. 0, not reported; RT, radiotherapy; HT, hormone therapy.

ratio ¼ 1.37; 95% confidence interval 1.14e1.64) and locally advanced disease (hazard ratio ¼ 2.44; 95% confidence interval 1.83e3.25), this advantage was statistically significantly greater in the latter than in the former. Radiotherapy Alone versus Neoadjuvant Hormone Therapy and Radiotherapy Jones et al. [20] defined low-risk disease as a Gleason score 6, PSA  10 ng/ml and a clinical stage  T2a; intermediate-risk disease as a Gleason score of 7 or a Gleason score 6 with a PSA > 10 and 30%, platelet count > 100  103/ ml, life expectancy  10 years (excluding death from prostate cancer) at study entry, negative bone scan and pelvic lymph node assessment using MRI or CT within 6 months of randomisation. Low-risk patients with radiographic evidence using endorectal coil MRI of extracapsular extension or seminal vesicle invasion. Exclusion criteria: history of malignant disease apart from non-melanoma skin cancer, any history of HT. Length of follow-up (median): 8.2 (interquartile range 7e9.5) years (PSA recurrence), 7.6 (range 0.5e11) years (overall survival). No. in trial arm: RT: n ¼ 104; RT/HT: n ¼ 102 Age (median, range): RT: 73 (51e81) years; RT/HT: 72 (49e82) years. ECOG performance status: RT: 0 (n ¼ 101), 1 (n ¼ 3); RT/HT: 0 (n ¼ 95), 1 (n ¼ 7). Baseline PSA level (median, range; ng/ml): RT: 11 (0.9e40); RT/HT: 11 (1.3e36). Gleason total score: RT: 5 or 6 (n ¼ 27), 3 þ 4 (n ¼ 37), 4 þ 3 (n ¼ 24), 8e10 (n ¼ 16); RT/HT: 5 or 6 (n ¼ 30), 3 þ 4 (n ¼ 35), 4 þ 3 (n ¼ 23), 8e10 (n ¼ 14). Clinical T classification: RT: T1b (n ¼ 3), T1c (n ¼ 41), T2a (n ¼ 26), T2b (n ¼ 34); RT/HT: T1b (n ¼ 1), T1c (n ¼ 54), T2a (n ¼ 20), T2b (n ¼ 27). Treatment stratification: RT: PSA of 20e40 ng/ml (n ¼ 13), Gleason score  7 (n ¼ 64), PSA of 20e40 ng/ml and Gleason score  6 (n ¼ 24), Low risk and endorectal MRI category T3 (n ¼ 3); RT/HT: PSA of 20e40 ng/ml (n ¼ 12), Gleason score  7 (n ¼ 64), PSA of 20e40 ng/ml and Gleason score  6 (n ¼ 24), Low risk and endorectal MRI category T3 (n ¼ 2). RT alone versus RT þ androgen suppression therapy given neoadjuvantly (2 months), concurrently (2 months) and adjuvantly (2 months) (RT/HT) RT: Patients were treated once a day/5 days a week at a daily dose of 1.8 Gy for the initial 25 treatments and 2 Gy for the final 11 treatments totalling 70.35 Gy to the prostate (and seminal vesicles?) plus a 1.5 cm margin using a 4-field 3D-CRT technique. HT: Leuprolide acetate (n ¼ 88) was delivered intramuscularly each month at a dose of 7.5 mg or 22.5 mg every 3 months. 3.6 mg goserelin (n ¼ 10) was administered subcutaneously each month (or 10.8 mg every 3 months). Flutamide (n ¼ 98) every 8 h at a dose of 250 mg starting 1e3 days before leuprolide acetate/goserelin. Flutamide was discontinued if either aspartate aminotransferase or alanine aminotransferase exceeded 2  the upper limit of normal or the patient developed drug-induced diarrhoea or anaemia causing clinical symptoms. The treating physician assessed potency at randomisation. PSA recurrence, overall survival, cause-specific mortality, adverse events. The groups seem to be comparable at baseline, although not sure if the T stage is slightly higher in the RT patients compared with the RT/HT patients. Information on treatment adherence available for 103/104 RT patients and 98/102 RT/HT patients. RT: All patients had RT per protocol. All RT/HT patients completed 6 months of leuprolide/goserelin, but 27/98 did not complete 6 months of flutamide treatment due to adverse events. ITT analyses undertaken. (continued on next page)

e34 (continued)

M. Schmidt-Hansen et al. / Clinical Oncology (2014) e21ee46

Bias/quality item

Authors’ judgement (low/unclear/high risk of bias)

Support for judgement

Random sequence generation (selection bias) Allocation concealment (selection bias)

Low

Blinding of outcome assessment (detection bias): objective outcomes Blinding of outcome assessment (detection bias): subjective outcomes Incomplete outcome data (attrition bias)

Unclear

Central randomisation with stratification for baseline PSA level and Gleason score. A permuted blocks randomisation algorithm was used with a block size of 4. No information reported.

Unclear

No information reported.

Low

Selective reporting (reporting bias)

Unclear

The data from all the randomised patients have been analysed for all reported outcomes apart from adverse events where data from 103/ 104 RT patients and from 98/ 102 RT/HT patients included. Distant metastasis-free survival and locoregional control not reported

Low

Study: EORTC 22863 [7,8] Methods

Participants

Interventions

Outcomes

Study design: randomised controlled trial (multicentre phase III) Country: international Study period: 1987e1995 Inclusion criteria: age < 80 years, WHO performance status 0e2, newly diagnosed histologically proven T1e2 prostatic adenocarcinoma with WHO histological grade 3, or T3e4 prostatic adenocarcinoma of any histological grade. Exclusion criteria: history of malignant disease apart from adequately treated basal-cell carcinoma of the skin, or evidence of distant metastases, including involvement of common iliac or para-aortic lymph nodes. Length of follow-up (median and interquartile range): 9.1 (5.1e12.6) years. No. in trial arm: RT: n ¼ 208; RT/HT: n ¼ 207 Age (median, interquartile range): RT: 70 (65e75) years; RT/HT: 71 (67e75) years. WHO performance status: RT: 0 (n ¼ 164), 1 (n ¼ 38), 2 (n ¼ 4), not documented (n ¼ 2); RT/HT: 0 (n ¼ 162), 1 (n ¼ 37), 2 (n ¼ 7), not documented (n ¼ 1). WHO histopathological grade: RT: G1 (n ¼ 39), G2 (n ¼ 96), G3 (n ¼ 68), not documented (n ¼ 5); RT/HT: G1 (n ¼ 44), G2 (n ¼ 98), G3 (n ¼ 63), not documented (n ¼ 2). Gleason total score: RT: 2e4 (n ¼ 16), 5e6 (n ¼ 40), 7e10 (n ¼ 71), not documented (n ¼ 81); RT/HT: 2e4 (n ¼ 11), 5e6 (n ¼ 50), 7e10 (n ¼ 66), not documented (n ¼ 80). Clinical T classification: RT: T1 (n ¼ 2), T2 (n ¼ 20), T3 (n ¼ 167), T4 (n ¼ 18), not documented (n ¼ 1); RT/HT: T1 (n ¼ 2), T2 (n ¼ 18), T3 (n ¼ 167), T4 (n ¼ 20), not documented (n ¼ 0). N classification: RT: N0 (n ¼ 183), N1 (n ¼ 5), N2 (n ¼ 1), N4 (n ¼ 1), NX (n ¼ 18); RT/HT: N0 (n ¼ 184), N1 (n ¼ 4), N2 (n ¼ 5), N4 (n ¼ 0), NX (n ¼ 14). T according to grade (stratification): RT: T1e2 G3 (n ¼ 20), T3e4 any G (n ¼ 188); RT/HT: T1e2 G3 (n ¼ 20), T3e4 any G (n ¼ 187). Baseline PSA concentration: RT: < 4 mg/l (n ¼ 10), 4 to < 10 mg/l (n ¼ 23), 10 to < 20 mg/l (n ¼ 36), 20 to < 40 mg/l (n ¼ 49), >40 mg/l (n ¼ 67), not documented (n ¼ 23); RT/HT: < 4 mg/l (n ¼ 16), 4 to < 10 mg/l (n ¼ 24), 10 to < 20 mg/l (n ¼ 29), 20 to < 40 mg/l (n ¼ 47), >40 mg/l (n ¼ 72), not documented (n ¼ 19). Chronic disease: RT: None (n ¼ 100), cardiovascular (n ¼ 63), other (n ¼ 42), not documented (n ¼ 3); RT/HT: None (n ¼ 111), cardiovascular (n ¼ 53), other (n ¼ 43), not documented (n ¼ 0). RT alone versus RT þ immediate androgen suppression (RT/HT) RT (both groups): patients were treated once a day/5 days a week/7 weeks and consisted of planning target volume 1 (the whole pelvis irradiated up to 50 Gy) and planning target volume 2 (the prostate and the seminal vesicles irradiated with an additional 20 Gy). HT: 3.6 mg goserelin acetate administered subcutaneously every 4 weeks starting the first day of pelvic RT and continued for 3 years. Cyproterone acetate administered orally for 1 month, 50 mg three times a day starting a week before goserelin. Clinical disease-free survival, overall survival, distant metastasis-free survival, cause-specific mortality and locoregional control. Quality of life was not obtained in this study.

M. Schmidt-Hansen et al. / Clinical Oncology (2014) e21ee46

e35

(continued) Notes

The groups seem to be comparable at baseline. RT: 203/208 started treatment and 200/203 completed treatment; RT/HT: 198/207 started RT/HT treatment, 201/207 completed RT treatment, 150/198 completed HT treatment. ITT analyses undertaken.

Bias/quality item

Authors’ judgement (low/ unclear/high risk of bias)

Support for judgement

Random sequence generation (selection bias) Allocation concealment (selection bias)

Low

Blinding of outcome assessment (detection bias): objective outcomes Blinding of outcome assessment (detection bias): subjective outcomes Incomplete outcome data (attrition bias) Selective reporting (reporting bias)

Unclear

Central randomisation. Randomisation used the minimisation technique with institution, clinical stage (T1e2 WHO grade 3 versus T3e4 WHO grade 1e3), results of pelvic-lymph-node dissection (N0 versus N1), and irradiation fields extension (extended versus limited fields) as minimisation factors. No information reported apart from ‘Cause of death was prospectively documented by the treating physician and was not subjected to central independent review’.

Low

Unclear

Low Unclear

The data from all the randomised patients have been analysed for all reported outcomes. No adverse events reported

Study: EPC23-25 [9,10] e analysis of the patients who received RT as standard care Methods

Participants

Interventions

Outcomes Notes

Study design: randomised controlled trial (multicentre) Country: international Study period: ongoing Inclusion criteria: patients aged  18 (trials 23, 24) or 18e75 (trial 25) years with clinically or pathologically confirmed T1e2N0/Nx or T3e4 any N, or any T Nþ prostate cancer with no evidence of distant metastases. Exclusion criteria: prior systemic therapy for prostate cancer. Length of follow-up (median): 7.2 years. No. in trial arm: RT: n ¼ 671; RT/HT: n ¼ 699 Age (mean, range): RT: 69.3 (47e82) years; RT/HT: 69.6 (48e85) years. Gleason score: RT: 2e4 (22.1%), 5e6 (52.2%), 7e10 (25.6%), not known (0.1%); RT/HT: 2e4 (23.8%), 5e6 (49.6%), 7e10 (23.7%), not known (1.4%). T stage: RT: T1e2 (79.1%), T3 (20.1%), T4 (0.8%); RT/HT: T1e2 (77.5%), T3 (21.7%), T4 (0.7%). N stage: RT: N0 (33.1%), Nx (65.9%), Nþ (1%); RT/HT: N0 (30.8%), Nx (68.2%), Nþ (1%). Median PSA level (ng/ml; range): RT: before RT (11.2; 0.4e204), at randomisation (3.5; not quantifiablee147.2), localised disease (3.4; not quantifiablee101.3), locally advanced disease (4; not quantifiablee147.2); RT/HT: before RT (11.3; 0.3e681), at randomisation (3.4; not quantifiablee119.3), localised disease (3.3; not quantifiablee69), locally advanced disease (3.8; not quantifiablee119.3) Use of neoadjuvant therapy: RT: 32.5%; RT/HT: 30%. Trial: RT: 23 (47.7%), 24 (48.4%), 25 (3.9%); RT/HT: 23 (46.5%), 24 (47.9%), 25 (5.6%). RT followed by placebo once daily (RT) versus RT followed by oral biccalutamide 150 mg once daily for  2 years (RT/HT) RT (both groups): EPC programme was designed to reflect current standard care worldwide, therefore RT techniques and dose fractionation schedules as well as type and duration of HT were not specified in the protocol. Objective progression-free survival, overall survival, PSA and tolerability. Not sure if the baseline characteristics of RT/HT are better than those of RT. Information about the characteristics of the RT used was not collected, but the authors attempted to collect this information retrospectively in trials 24 and 25, but not 23. Some records are no longer available or incomplete. Of the 725 patients receiving RT in trials 24 and 25, information was collected for 681 patients on the type of RT given, 643 patients on the dose of RT, 619 patients on the number of fractions and 621 patients on the duration of therapy. Of these data, 93.4% of patients received external-beam RT alone (median dose 64 Gy, median fractions ¼ 32, median duration 6.6 weeks), whereas 6.5% received external-beam RT and brachytherapy. ITT analyses undertaken.

Bias/quality item

Authors’ judgement (low/ unclear/high risk of bias)

Support for judgement

Random sequence generation (selection bias) Allocation concealment (selection bias) Blinding of outcome assessment (detection bias): objective outcomes Blinding of outcome assessment (detection bias): subjective outcomes

Unclear Unclear Unclear

No information reported No information reported No information reported

Unclear

No information reported (continued on next page)

e36 (continued)

M. Schmidt-Hansen et al. / Clinical Oncology (2014) e21ee46

Incomplete outcome data (attrition bias)

Low

Selective reporting (reporting bias)

Low

All data appear to have been included in the analyses The main relevant outcomes appear to have been reported.

Study: Fellows et al. [36] Methods

Participants

Interventions

Outcomes Notes

Study design: randomised controlled trial Country: Great Britain Study period: 1980e1985 Inclusion criteria: patients with histologically proven prostate cancer, no previous treatment with HT or RT, no evidence of metastases on chest X-ray or isotope bone scan. Exclusion criteria: patients with serum acid phosphatise > upper limit of normal for the local laboratory. Length of follow-up (range of medians for the treatment groups): 4e5.2 years. No. in trial arm: RT: n ¼ 88; HT: n ¼ 90; HT/RT: n ¼ 99. Age (years): RT: 20, or Gleason > 7, or T2ceT4) (n ¼ 222); HT3/ RT: intermediate (n ¼ 39), high (n ¼ 226); HT6/RT: intermediate (n ¼ 43), high (n ¼ 224). RT alone versus HT for 3 months followed by RT (HT3/RT) versus HT for 6 months followed by RT (HT6/RT) RT: 2 Gy a day/5 days a week/6.5e7 weeks to a dose of 66 Gy to the prostate and the seminal vesicles. HT: 3.6 mg goserelin acetate administered subcutaneously every month and 250 mg flutamide given orally 3 times a day. HT started 2 months before radiation in group HT3/RT and 5 months before radiation in group HT6/RT. Overall survival, distant progression, prostate cancer-specific mortality, secondary therapeutic intervention, event-free survival, local progression. The groups seem to be comparable at baseline. RT: 268/276 received RT; HT3/RT: 264/270 received HT3; HT6/RT: 266/272 received HT6. Paper states that ITT analyses undertaken.

Bias/quality item

Authors’ judgement (low/unclear/high risk of bias)

Support for judgement

Random sequence generation (selection bias) Allocation concealment (selection bias) Blinding of outcome assessment (detection bias): objective outcomes Blinding of outcome assessment (detection bias): subjective outcomes Incomplete outcome data (attrition bias)

Low

Unclear

‘Patients were randomised by the minimisation technique at the TROG Central Trials Office in Newcastle’. No information reported.

Unclear

No information reported.

Low

RT: 270/276 analysed; HT3/RT: 265/270 analysed; HT6/RT: 267/272 analysed. No adverse events reported

Selective reporting (reporting bias)

Low

Unclear

e44

M. Schmidt-Hansen et al. / Clinical Oncology (2014) e21ee46

Study: Zagars et al. [14] Methods

Participants

Interventions

Outcomes Notes

Study design: randomised controlled trial Country: USA Study period: 1967e1973 Inclusion criteria: clinical stage C adenocarcinoma of the prostate. No patient had received any prior treatment for his prostatic carcinoma, except in cases where transurethral resection of the prostate was carried out. Exclusion criteria: none listed. Length of follow-up: unclear, but seems to be up to 15 years. No. in trial arm: RT: n ¼ 43; RT/HT: n ¼ 39 Age: all patients: mean and median ¼ 64 (range 51e73) years. Prostatism: RT: n ¼ 26; RT/HT: n ¼ 29. TURP: RT: n ¼ 21; RT/HT: n ¼ 21. Grade: RT: 1 (n ¼ 15), 2e3 (n ¼ 11), 4 (n ¼ 2); RT/HT: 1 (n ¼ 11), 2e3 (n ¼ 2), 4 (n ¼ 2). Bladder invasion: RT: n ¼ 2; RT/HT: n ¼ 2. Sidewall involved: RT: n ¼ 7; RT/HT: n ¼ 14. Elevated serum prostatic acid phosphatase: RT: n ¼ 6; RT/HT: n ¼ 5. Creatine >1.5 mg/dl: RT: n ¼ 1; RT/HT: n ¼ 1. Hydronephrosis: RT: n ¼ 3; RT/HT: n ¼ 0. RT alone versus RT þ immediate oestrogen (RT/HT) RT: patients were treated with high-energy (18-25 MV) photon beams via a four-field portal arrangement using app 10  10 cm anteroposterior and 10  8 cm lateral fields. The inferior margin of these fields was at or just cranial to the lower border of the ischial tuberosities, and the posterior border of the lateral fields bisected the rectal lumen. After 50 Gy was delivered in 5 weeks at 2 Gy per fraction with this technique, a reduced volume of 8  8 or 9  9 cm was given an additional 20 Gy in 2 weeks through anteroposterior parallel-opposed fields. HT: 5 mg daily diethylstilbestrol started immediately after completion of RT. In 1972 dose reduced to 2 mg daily. HT to continue indefinitely. Clinical disease-free survival, overall survival, distant metastasis-free survival, cause-specific mortality and locoregional control. Quality of life was not obtained in this study. The groups seem to be comparable at baseline. RT: 40/43 were evaluable (n ¼ 1 did not have documented carcinoma, n ¼ 1 refused to complete RT and n ¼ 1 was lost to follow-up); RT/HT: 38/39 were evaluable (n ¼ 1 had bone metastasis). 4/38 did not receive HT, 20/34 patients treated with HT received 5 mg daily, 12/34 received 2 mg daily and 2/34 received 12 mg daily chlorotrianisene (Tace), 12/34 patients terminated HT prematurely (i.e. after relapse or death) after taking HT for 9e93 (mean ¼ 42, median ¼ 41) months. The reasons for discontinuing HT were unclear (n ¼ 5), cardiac or cerebrovascular problems (n ¼ 5), refusal of further treatment due to feminising effects (n ¼ 2) Treatment after relapse not specified in protocol: 27/40 RT patients relapsed, 14 of whom received oestrogen only, 9 received orchiecyomy and oestrogen, 3 had orchiectomy and 1 received non-hormonal treatment. 11/38 HT/RT patients relapsed: 6 underwent orchiectomy, 1 received higher dose of oestrogen, 1 underwent transurethral resection of the prostate only, 1 received chemotherapy and 2 had no specific additional treatment. ITT and per-protocol analyses undertaken. Authors report that results did not differ between these two methods op analysis.

Bias/quality item

Authors’ judgement (low/unclear/high risk of bias)

Support for judgement

Random sequence generation (selection bias) Allocation concealment (selection bias) Blinding of outcome assessment (detection bias): objective outcomes Blinding of outcome assessment (detection bias): subjective outcomes Incomplete outcome data (attrition bias)

Unclear

No details reported.

Unclear

No details reported.

Unclear

No details reported.

Unclear

No details reported.

Low

Selective reporting (reporting bias)

Unclear

The data from all the randomised patients have been analysed for all reported outcomes. No adverse events of RT reported

ADT, androgen deprivation therapy; LHRH, luteinising hormone releasing hormone; PSA, prostate-specific antigen; HT, hormone therapy; RT, radiotherapy; MRI, magnetic resonance imaging; CT, computed tomography; ITT, intention to treat.

M. Schmidt-Hansen et al. / Clinical Oncology (2014) e21ee46

References [1] Wo JY, Zietman AL. Why does androgen deprivation enhance the results of radiation therapy? Urol Oncol 2008;26:522e529. [2] Alonzi R, Padhani AR, Taylor J, et al. Antivascular effects of neoadjuvant androgen deprivation for prostate cancer: an in vivo human study using susceptibility and relaxivity dynamic MRI. Int J Radiat Oncol Biol Phys 2011;80:721e727. [3] Kumar S, Shelley M, Harrison C, et al. Neo-adjuvant and adjuvant hormone therapy for localised and locally advanced prostate cancer. Cochrane Database Sys Rev 2006;(4). http:// dx.doi.org/10.1002/14651858 CD006019 pub2. Art no. CD006019. [4] Higgins JPT, Green SE. Cochrane Handbook for Systematic Reviews of Inverventions version 5.1.0 [updated March 2011]. The Cochrane Collaboration; 2011. [5] Review Manager (RevMan) [Computer program].Version 5.2. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration; 2012. [6] Tierney JF, Stewart LA, Ghersi D, et al. Practical methods for incorporating summary time-to-event data into meta-analysis. Trials 2007;8:1e16. [7] Bolla M. Three years of adjuvant androgen deprivation with goserelin in patients with locally advanced prostate cancer treated with radiotherapy: results at 10 years of EORTC trial 22863. Eur J Cancer 2009. Conference Suppl.:2e3. [8] Bolla M, Van TG, Warde P, et al. External irradiation with or without long-term androgen suppression for prostate cancer with high metastatic risk: 10-year results of an EORTC randomised study. Lancet Oncol 2010;11:1066e1073. [9] See WA, Tyrrell CJ, Casodex Early Prostate Cancer Trialists’ Group. The addition of bicalutamide 150 mg to radiotherapy significantly improves overall survival in men with locally advanced prostate cancer. J Cancer Res Clin Oncol 2006;132(Suppl. 1):S7eS16. [10] Tyrrell CJ, Payne H, See WA, et al. Bicalutamide (‘Casodex’) 150 mg as adjuvant to radiotherapy in patients with localised or locally advanced prostate cancer: results from the randomised Early Prostate Cancer Programme. Radiother Oncol 2005;76:4e10. [11] Efstathiou JA, Bae K, Shipley WU, et al. Obesity and mortality in men with locally advanced prostate cancer: analysis of RTOG 85-31. Cancer 2007;110:2691e2699. [12] Efstathiou JA, Bae K, Shipley WU, et al. Cardiovascular mortality after androgen deprivation therapy for locally advanced prostate cancer: RTOG 85-31. J Clin Oncol 2009;27:92e99. [13] Pilepich MV, Winter K, Lawton CA, et al. Androgen suppression adjuvant to definitive radiotherapy in prostate carcinoma e long-term results of phase III RTOG 85-31. Int J Radiat Oncol Biol Phys 2005;61:1285e1290. [14] Zagars GK, Johnson DE, von Eschenbach AC, et al. Adjuvant estrogen following radiation therapy for stage C adenocarcinoma of the prostate: long-term results of a prospective randomized study. Int J Radiat Oncol Biol Phys 1988;14:1085e1091. [15] Pilepich MV, Winter K, John MJ, et al. Phase III Radiation Therapy Oncology Group (RTOG) trial 86-10 of androgen deprivation adjuvant to definitive radiotherapy in locally advanced carcinoma of the prostate. Int J Radiat Oncol Biol Phys 2001;50:1243e1252. [16] Roach M, Bae K, Speight J, et al. Short-term neoadjuvant androgen deprivation therapy and external-beam radiotherapy for locally advanced prostate cancer: long-term results of RTOG 8610. J Clin Oncol 2008;26:585e591.

e45

[17] Denham JW, Steigler A, Lamb DS, et al. Short-term neoadjuvant androgen deprivation and radiotherapy for locally advanced prostate cancer: 10-year data from the TROG 96.01 randomised trial. Lancet Oncol 2011;12:451e459. [18] Wilcox C, Kautto A, Steigler A, et al. Androgen deprivation therapy for prostate cancer does not increase cardiovascular mortality in the long term. Oncology 2012;82:56e58. [19] Laverdiere J, Nabid A, De Bedoya LD, et al. The efficacy and sequencing of a short course of androgen suppression on freedom from biochemical failure when administered with radiation therapy for T2-T3 prostate cancer. J Urol 2004;171:1137e1140. [20] Jones CU, Hunt D, McGowan DG, et al. Radiotherapy and short-term androgen deprivation for localized prostate cancer. N Engl J Med 2011;365:107e118. [21] Granfors T, Modig H, Damber JE, et al. Combined orchiectomy and external radiotherapy versus radiotherapy alone for nonmetastatic prostate cancer with or without pelvic lymph node involvement: a prospective randomized study. J Urol 1998;159:2030e2034. [22] Granfors T, Modig H, Damber JE, et al. Long-term followup of a randomized study of locally advanced prostate cancer treated with combined orchiectomy and external radiotherapy versus radiotherapy alone. J Urol 2006;176:544e547. [23] D’Amico AV, Chen MH, Renshaw AA, et al. Androgen suppression and radiation vs radiation alone for prostate cancer: a randomized trial. JAMA 2008;299:289e295. [24] D’Amico AV, Chen MH, Renshaw AA, et al. Risk of prostate cancer recurrence in men treated with radiation alone or in conjunction with combined or less than combined androgen suppression therapy. J Clin Oncol 2008;26:2979e2983. [25] D’Amico AV, Manola JM, Loffredo M, et al. 6-month androgen suppression plus radiation therapy vs radiation therapy alone for patients with clinically localized prostate cancer: a randomized controlled trial. JAMA 2004;292:821e827. [26] Nguyen PL, Chen MH, Beard CJ, et al. Radiation with or without 6 months of androgen suppression therapy in intermediate- and high-risk clinically localized prostate cancer: a postrandomization analysis by risk group. Int J Radiat Oncol Biol Phys 2010;77:1046e1052. [27] Nguyen PL, Chen MH, Renshaw AA, et al. Survival following radiation and androgen suppression therapy for prostate cancer in healthy older men: implications for screening recommendations. Int J Radiat Oncol Biol Phys 2010;76:337e341. [28] Mottet N, Peneau M, Mazeron JJ, et al. Addition of radiotherapy to long-term androgen deprivation in locally advanced prostate cancer: an open randomised phase 3 trial. Eur Urol 2012;62:213e219. [29] Mottet N. Radiotherapy combined with androgen deprivation vs androgen deprivation alone in clinically locally advanced prostate cancer (PCA) T3-T4,N0,M0 in a multicenter randomised phase III study. J Urol 2010. Conference:4. [30] Warde P, Tsuji D, Bristow R. A randomized phase III study of neoadjuvant hormonal therapy in patients with localized prostate cancer. Clin Genitourin Cancer 2006;5:235e237. [31] Warde P, Mason M, Ding K, et al. Combined androgen deprivation therapy and radiation therapy for locally advanced prostate cancer: a randomised, phase 3 trial. Lancet 2011;378:2104e2111. [32] Warde PR, Mason MD, Sydes MR, et al. Intergroup randomized phase III study of androgen deprivation therapy (ADT) plus radiation therapy (RT) in locally advanced prostate cancer (CaP) (NCIC-CTG, SWOG, MRC-UK, INT: T94-0110; NCT00002633). J Clin Oncol 2010;28:959.

e46

M. Schmidt-Hansen et al. / Clinical Oncology (2014) e21ee46

[33] Gospodarowicz MK. Final analysis of intergroup randomized phase III study of androgen deprivation therapy (ADT) radiation therapy (RT) in locally advanced prostate cancer (NCICCTG, Swog, MRC-UK, INT: T94e0110; NCT00002633). Int J Radiat Oncol Biol Phys 2012. Conference:3. [34] Fransson P, Lund JA, Damber JE, et al. Quality of life in patients with locally advanced prostate cancer given endocrine treatment with or without radiotherapy: 4-year follow-up of SPCG-7/SFUO-3, an open-label, randomised, phase III trial. Lancet Oncol 2009;10:370e380. [35] Widmark A, Klepp O, Solberg A, et al. Endocrine treatment, with or without radiotherapy, in locally advanced prostate cancer (SPCG-7/SFUO-3): an open randomised phase III trial. Lancet 2009;373:301e308. Erratum 2009;373:1174. [36] Fellows GJ, Clark PB, Beynon LL, et al. Treatment of advanced localised prostatic cancer by orchiectomy, radiotherapy, or combined treatment. A Medical Research Council study.

[37] [38]

[39]

[40]

[41]

Urological Cancer Working Party e Subgroup on Prostatic Cancer. Br J Urol 1992;70:304e309. National Institute for Health and Care Excellence (NICE). Prostate cancer: diagnosis and treatment, www.nice.org.uk; 2014. Williams S, Buyyounouski M, Kestin L, et al. Predictors of androgen deprivation therapy efficacy combined with prostatic irradiation: the central role of tumour stage and radiation dose. Int J Radiat Oncol Biol Phys 2011;79:724e731. Nabid A, Carrier N, Martin A-G, et al. Duration of androgen deprivation therapy in high-risk prostate cancer: a randomized trial. J Clin Oncol 2013;31. abstract LBA4510. Bolla M, Maingon P, van den Bergh F, et al. 3D-CRT/IMRT with/ without short term androgen deprivation in localized T1BCT2AN0M0 prostatic carcinoma (EORTC 22991). ESTRO 33 Congress report 2014. p. 12. Zelen M. The randomization and stratification of patients to clinical trials. J Chronic Dis 1974;27:365e375.

Hormone and radiotherapy versus hormone or radiotherapy alone for non-metastatic prostate cancer: a systematic review with meta-analyses.

Radiotherapy is standard treatment for localised prostate cancer and is often combined with hormone treatment to prevent androgen stimulation of prost...
2MB Sizes 0 Downloads 3 Views