Abstract
Purpose
Assess the impact of PSA-complete response (PSA-CR), measured at the end of external beam radiotherapy (EBRT) and short-term hormonal therapy (STHT), on treatment outcomes.
Design
The Phase III RTOG-9413 trial had as part of its original protocol the assessment of PSA-CR, i.e. PSA≤0.3ng/ml, at the end of STHT as a secondary endpoint. STHT consisted of flutamide plus an LHRH-agonist for 4 months. Kaplan-Meier method was used to estimate overall survival (OS) and disease-free survival (DFS). Cumulative incidence was used to estimate biochemical failure (BF), distant metastasis (DM), and disease-specific survival (DSS). Univariate and multivariate analyses were performed to correlate PSA-CR after STHT with all endpoints, and the following variables were considered for analysis: PSA at baseline, Gleason score, treatment arm, age, and baseline testosterone. Phoenix-consensus was used to define PSA failure.
Results
For 1070 evaluable patients, the median PSA at the end of STHT was 0.2ng/mL. A total of 744 patients (70%) had PSA-CR. With median follow-up of 7.2 years, failure to obtain PSA-CR was significantly associated with worse DSS (p=0.0003; hazard ratio, 2.03[95%CI, 1.38–2.97]) and DFS (p=0.003; 1.28[1.09–1.50]), as well as with a higher incidence of DM (p=0.0002; 1.92[1.37–2.69]) and BF (p<0.0001; 1.57[1.29–1.91]). The other factors associated with worse DSS were Gleason score 8–10 (p=0.0002; 3.06[1.71–5.47]) and PSA>20ng/mL (p=0.04; 1.55[1.02–2.30]).
Conclusion
Failure to obtain a post STHT and EBRT PSA-CR (≤0.3ng/mL) appears to be an independent predictor of unfavorable outcomes, and may help identify patients who could benefit from the addition of long-term androgen ablation.
INTRODUCTION
Androgen deprivation therapy (ADT) in combination with radiation therapy is frequently used in the treatment of patients with locally advanced prostate cancer1. ADT duration is usually predetermined based on tumor characteristics, mainly Gleason score, baseline prostatic specific antigen (PSA) and T stage (TNM staging system), on physicians’ experience and also patients’ characteristics (age, co-morbidities, life expectancy). Optimal ADT duration is unknown and patients are treated based on knowledge learned from clinical trials, most demonstrating that longer treatment duration for higher risk disease provides better treatment outcomes. ADT, however, is not innocuous and is associated with side effects in both the short- and long-term2. Therefore, the ideal balance between oncologic need and normal physiologic tolerance must be evaluated on an individual basis.
PSA and its various constructs, such as PSA doubling time, free PSA, PSA density, PSA nadir, time to PSA nadir, among others, are tools used for screening and post-treatment follow-up of prostate cancer, independent of the therapeutic option chosen3,4. However, most measurements of treatment efficacy are performed retrospectively and based on consecutive PSA measurements, not allowing for prospective interventions during treatment and, consequently, treatment individualization.
RTOG 94135 was a phase III randomized controlled trial that tested 2 different hypotheses in men with localized prostate cancer. First, whether total androgen suppression (TAS) plus whole pelvic radiation therapy improves outcomes when compared to TAS plus prostate only radiation therapy, and second, if neoadjuvant and concurrent TAS plus radiation therapy improves outcomes compared to adjuvant TAS plus radiation therapy. The original study protocol6 had in its design a planned assessment of “PSA complete response” (PSA-CR), i.e. PSA ≤0.3 ng/mL, at the discontinuation of hormonal therapy, to measure treatment efficacy. For this specific purposed, the protocol required a blood test with serum PSA for all patients at completion of hormonal therapy. This information was acquired prospectively for patients enrolled in this trial. The objective of this study is to investigate the correlation between PSA ≤0.3 ng/mL with treatment outcomes considered as endpoints on RTOG 9413 study. The hypothesis of this analysis is that patients with PSA-CR at the end of hormonal therapy will have better treatment outcomes than those with a PSA nadir higher than 0.3 ng/mL at the same time point.
PATIENTS AND METHODS
RTOG 9413 trial accrued men with histologically confirmed adenocarcinoma of the prostate, who had an estimated risk of lymph node involvement higher than 15% but with negative nodes, and PSA ≤ 100 ng/ml. A total of 1279 patients were enrolled and treated between April, 1995 and June, 1999 and were randomized to one of four arms, receiving prostate only radiation therapy (PORT) or whole pelvis radiation therapy (WPRT), which was followed by a boost to the prostate only, in combination with neoadjuvant and concomitant hormonal therapy (NCHT) or adjuvant hormonal therapy (AHT). All patients were planned to receive a total dose of 70.2 Gy in 39 fractions, with the difference that patients receiving WPRT were treated with 2 plans, one for the whole pelvis to 50.4 Gy and a second to the prostate only reaching the total dose, while patients receiving PORT were treated with a single plan to the prostate and seminal vesicles. In regards to HT, all patients received total androgen suppression (TAS) for a total period of 4 months, which consisted of monthly leuprolide acetate 7.5 mg intramuscularly or goserelin acetate 3.6 mg subcutaneously, and fflutamide 250 mg tid orally. Patients receiving NCHT received their first LHRH-agonist injection 2 months before RT and continued HT for the duration of RT, while those receiving AHT were started on HT immediately after completion of RT.
Patients eligible for the current analysis were those treated on the RTOG 9413 protocol that had available baseline testosterone, end of hormonal therapy date and PSA data at the end of hormonal therapy, defined as a PSA evaluation done within 120 days following the completion of hormone therapy. Outcome parameters assessed on this study were biochemical failure (BF), local progression (LP), distant metastasis (DM), disease-free survival (DFS), disease specific survival (DSS) and overall survival (OS). RTOG-ASTRO Phoenix consensus definition7 was used to characterize biochemical failure.
For endpoints OS and DFS, the Kaplan-Meier method8 was used to estimate yearly survival rates. For all other endpoints BF, LP, DM and DSS, the cumulative incidence approach was used to estimate yearly incidence rates9 and Gray’s test10 was used to test for differences. Univariate analysis was performed to correlate PSA-CR (PSA ≤0.3 ng/ml) after TAS with all endpoints, and the following variables were considered for multivariate analysis: PSA at baseline, Gleason score, treatment arm, age, and baseline testosterone. The Cox proportional hazards regression model11 was used for univariate and multivariate analyses.
RESULTS
A total of 1279 patients were enrolled in the RTOG 9413 protocol, of which 1070 patients were eligible for the current study. Reasons for exclusion included a combination of absence of baseline testosterone (n=82), no end of hormonal therapy date (n=44) or no PSA data at the end of hormonal therapy (n=144). Patients’ characteristics are summarized on Table 1. A balance in pre-treatment characteristics, including age, baseline PSA, T stage, Gleason score, baseline testosterone, between the 4 treatment arms is noticed after the exclusion of 209 patients non-eligible for the present study. Median PSA at the end of hormonal therapy for all patients was 0.2 ng/mL (range: 0 - 96.2). A total of 744 patients (70%) reached a PSA-CR at the end of hormonal therapy. The time to reach PSA nadir was 3.7 months, ranging from 0.8 to 8.2 months. With a median follow-up of 7.2 years, higher rates of BF (56% vs. 38%), LP (15% vs. 10%), DM (20% vs. 10%) and worse DFS (28% vs. 40%), DSS (85% vs. 93%) and OS (67% vs. 69%) were observed in those 326 patients who failed to achieve PSA-CR. Univariate analysis showed that 7-year BF, DFS, DM and DSS rates were statistically significant better for patients with end of HT PSA ≤0.3 ng/mL, while LP and OS rates were, not statistically significant between both groups (Table 2).
TABLE 1.
Patients’ characteristics.
| Hormones +
RT Whole Pelvis+Boost(n=280) |
Hormones +
RT Prostate Alone(n=269) |
RT Whole Pelvis+Boost + Hormones(n=260) | RT Prostate Alone + Hormones(n=261) | |||||
|---|---|---|---|---|---|---|---|---|
| Age (year) | ||||||||
| Median | 70 | 70 | 70 | 70 | ||||
| Range | 46–83 | 44–87 | 50–84 | 45–85 | ||||
| PSA (ng/mL) | ||||||||
| Median | 23.1 | 22.9 | 22.1 | 23.9 | ||||
| Range | 2.9–97.6 | 4.1–95.1 | 3.0–98.0 | 4.1–98.2 | ||||
| n | % | n | % | n | % | n | % | |
|
|
||||||||
| T-Stage | ||||||||
| T1c,T2a | 66 | 24 | 68 | 25 | 52 | 20 | 56 | 21 |
| T1b,T2b | 29 | 10 | 30 | 11 | 20 | 8 | 33 | 13 |
| T2c-T4 | 185 | 66 | 171 | 64 | 188 | 72 | 172 | 66 |
| Baseline PSA | ||||||||
| <30 | 190 | 68 | 183 | 68 | 169 | 65 | 179 | 69 |
| ≥30 | 90 | 32 | 86 | 32 | 91 | 35 | 82 | 31 |
| ≥20 | 116 | 41 | 116 | 43 | 113 | 43 | 111 | 43 |
| >20 | 164 | 59 | 153 | 57 | 147 | 57 | 150 | 57 |
| ≤10 | 48 | 17 | 53 | 20 | 47 | 18 | 59 | 23 |
| >10 | 232 | 83 | 216 | 80 | 213 | 82 | 202 | 77 |
| Gleason (Institutional) | ||||||||
| 2–6 | 71 | 25 | 74 | 28 | 70 | 27 | 69 | 26 |
| 7 | 130 | 46 | 124 | 46 | 111 | 43 | 108 | 41 |
| 8–10 | 79 | 28 | 71 | 26 | 79 | 30 | 84 | 32 |
| Karnofsky status | ||||||||
| 70 | 4 | 1 | 3 | 1 | 3 | 1 | 6 | 2 |
| 80 | 22 | 8 | 15 | 6 | 16 | 6 | 16 | 6 |
| 90 | 123 | 44 | 122 | 45 | 107 | 41 | 107 | 41 |
| 100 | 131 | 47 | 129 | 48 | 134 | 52 | 132 | 51 |
| Race | ||||||||
| White | 202 | 72 | 191 | 71 | 193 | 74 | 183 | 70 |
| African American | 63 | 23 | 67 | 25 | 54 | 21 | 67 | 26 |
| Hispanic or Latino | 7 | 3 | 4 | 1 | 9 | 3 | 8 | 3 |
| Other | 7 | 3 | 4 | 1 | 4 | 2 | 2 | 1 |
| Unknown | 1 | <1 | 3 | 1 | 0 | 0 | 1 | <1 |
| Baseline | ||||||||
| Testosterone Status | ||||||||
| Normal | 236 | 84 | 215 | 80 | 214 | 82 | 204 | 78 |
| Abnormal | 44 | 16 | 54 | 20 | 46 | 18 | 57 | 22 |
TABLE 2.
Univariate analysis and survival rates at 7 years, comparing PSA ≤0.3 ng.mL with >0.3 mg/mL at the end of hormonal therapy.
| Outcome | Variable | Rates (7 yr) | HR | p-value |
|---|---|---|---|---|
| PSA Failure | ≤0.3 vs. | 38.1% vs. | - | |
| >0.3 | 56.4% | 1.83 (1.51, 2.20) | <0.0001 | |
| Disease-Free | ≤0.3 vs. | 40.1% vs. | - | |
| Survival | >0.3 | 28.1% | 1.44 (1.23, 1.69) | <0.0001 |
| Local Progression | ≤0.3 vs. | 10.4% vs. | - | |
| >0.3 | 14.7% | 1.41 (0.99, 2.00) | 0.06 | |
| Distant Metastasis | ≤0.3 vs. | 10.1% vs. | - | |
| >0.3 | 19.6% | 2.06 (1.49, 2.87) | <0.0001 | |
| Disease-Specific | ≤0.3 vs. | 93.1% vs. | - | |
| Survival | >0.3 | 85.4% | 2.24 (1.53, 3.29) | <0.0001 |
| Overall Survival | ≤0.3 vs. | 69.3% vs | - | |
| >0.3 | 66.6% | 1.08 (0.86, 1.35) | 0.52 |
On multivariate analysis, shown on Table 3, failure to reach a PSA-CR at the end of ADT was an independent predictor for prediction of BF, DFS, DM and DSS. Other factors significantly and independently associated with worse DSS were baseline PSA > 20 ng/mL and Gleason score 8–10. The only factor associated with poorer OS was baseline PSA >20 ng/mL.
Table 3.
Multivariate analysis of outcome parameters studied for all patients.
| Stratified Variables | Variable Categories | PSA Failure (Phoenix) | Disease-Free Survival | Local Progression | Distant Progression | Disease-specific Survival | Overall Survival | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
| |||||||||||||
| HR (95% CI) | p-value | HR (95% CI) | p-value | HR (95% CI) | p-value | HR (95% CI) | p-value | HR (95% CI) | p-value | HR (95% CI) | p-value | ||
|
| |||||||||||||
| PSA at End of HT | = 0.3 vs. | - | - | - | - | - | - | ||||||
| > 0.3 | 1.6 (1.3–1.9) | <0.001 | 1.3 (1.1–1.5) | 0.003 | 1.3 (0.9–1.8) | 0.20 | 1.9 (1.4–2.7) | <0.001 | 2.0 (1.4–3.0) | <0.001 | 1.1 (0.9–1.4) | 0.43 | |
|
| |||||||||||||
| PSA at Baseline | = 20 vs | - | - | - | - | - | - | ||||||
| .> 20 | 1.7 (1.4–2.1) | <0.001 | 1.5 (1.3–1.8) | <0.001 | 1.2 (0.9–1.8) | 0.22 | 1.2 (0.8–1.7) | 0.24 | 1.5 (1.0–2.3) | 0.04 | 1.3 (1.1–1.6) | 0.02 | |
|
| |||||||||||||
| Gleason | 2–6 vs. | - | - | - | - | - | - | ||||||
| 7 vs. | 1.5 (1.2–1.9) | <0.001 | 1.2 (1.0–1.5) | 0.03 | 1.6 (1.0–2.6) | 0.04 | 1.6 (1.0–2.6) | 0.04 | 1.8 (0.9–3.2) | 0.06 | 0.9 (0.7–1.2) | 0.44 | |
| 8–10 | 1.8 (1.4–2.4) | <0.001 | 1.3 (1.0–1.6) | 0.02 | 1.6 (0.9–2.6) | 0.06 | 2.2 (1.3–3.6) | 0.002 | 3.1 (1.7–5.5) | <0.001 | 1.2 (0.9–1.6) | 0.18 | |
|
| |||||||||||||
| Treatment Arm | Arm 1 vs. | - | - | - | - | - | - | ||||||
| Arm 2 vs. | 1.4 (1.1–1.8) | 0.01 | 1.3 (1.0–1.6) | 0.03 | 0.9 (0.6–1.6) | 0.95 | 1.2 (0.7–1.9) | 0.53 | 1.2 (0.7–2.2) | 0.46 | 0.9 (0.7–1.3) | 0.80 | |
| Arm 3 vs. | 1.2 (0.9–1.6) | 0.18 | 1.3 (1.1–1.6) | 0.008 | 1.1 (0.7–1.8) | 0.59 | 1.4 (0.8–2.2) | 0.17 | 1.4 (0.8–2.4) | 0.27 | 1.3 (0.9–1.7) | 0.09 | |
| Arm 4 | 1.1 (0.8–1.4) | 0.56 | 1.0 (0.8–1.3) | 0.73 | 0.8 (0.5–1.3) | 0.37 | 1.1 (0.7–1.8) | 0.59 | 1.2 (0.7–2.1) | 0.57 | 1.0 (0.7–1.4) | 0.81 | |
|
| |||||||||||||
| Age | Cont. | 0.9 (0.9–0.9) | <0.001 | 1.0 (0.9–1.1) | 0.79 | 0.9 (0.9–0.9) | 0.04 | 0.9 (0.9–1.0) | 0.66 | 1.0 (0.9–1.1) | 0.87 | 1.0 (1.0–1.1) | <0.001 |
Assessing separately the 326 patients (30.4%) who failed to reach PSA-CR at the end of HT, it was observed that a total of 169 patients presented PSA-CR at a later time (“late responders”), while the remaining 157 patients never reached PSA nadir below 0.3 ng/mL (“non-responders”). The median time to reach PSA ≤0.3 ng/mL for late complete responders was 7.2 months. When late responders are compared to non-responders, significant lower DM (p=0.02) and better DSS (p= 0.002) and OS (p=0.0001) were observed in patients with “late” PSA-CR. However, when late responders are compared to those patients who achieved PSA-CR at the end of HT, early complete response confers significantly lower BF (p<0.0001) and higher DFS (p=0.002).
On a post-hoc analysis of all patients who reached PSA-CR ≤0.3 ng/dL, independent of time to response, univariate analysis showed that 7-year BF, DFS, DM, DSS and now OS rates were statistically significant better for patients with end of HT PSA ≤0.3 ng/mL.
During the follow-up period a total of 357 patients died, being 104 patients due to prostate cancer. Among them, 55 patients (53%) achieved PSA-CR at the end of HT. Among the 253 patients who died due to other causes and 712 patients who remain alive, PSA-CR at the end of HT was noticed in 75% and 70% of patients, respectively. In an attempt to identify a PSA value at the end of hormone therapy that would predict risk of death, regression analysis showed that PSA value above 1.0 ng/mL might be associated with worse overall and cause specific survival. For 103 patients, multivariate analysis showed that PSA ≥ 1.0 ng/mL at the end of HT (p<0.0001), PSA at baseline higher than 20 ng/mL (p=0.02) and Gleason score 8 or higher (p=0.0001) were associated with worse DSS, while PSA ≥ 1.0 ng/mL at the end of HT (p=0.0008), baseline PSA (p=0.04) and age (p<0.0001) were associated with worse OS.
DISCUSSION
Phase III trials including intermediate- and high-risk prostate cancer patients treated with EBRT combined with ADT have demonstrated lower rates of distant metastasis, with improved biochemical control, cause specific survival and overall survival12–15. Although optimal duration of hormonal therapy is constantly under investigation in clinical trials, ADT for 28 to 36 months is generally accepted as standard for patients with high-risk prostate cancer14,15. Given the significant morbidity with 2–3 years of ADT, it is of interest to determine if the same beneficial effect can be observed with shorter courses of ADT and which patient population would benefit from this potential shortening, given that high-risk prostate cancer is a broad group of cancers classed under the same risk stratification that may possibly need different treatment approaches, such as duration of hormonal therapy.
Many prognostic and predictive factors have been evaluated for disease control and survival after initial therapy for prostate cancer, including initial PSA and Gleason score16, time to undetectable PSA17, PSA nadir at specific time points18–20 among others, but none has been tested by means of a clinical trial and translated to clinical practice. Currently, patients with prostate cancer have the duration of ADT pre-determined and based exclusively on baseline tumor characteristics, instead of individual responses to therapy. There are no means to differentiate patients who require shorter duration ADT from those who need longer periods of ADT, nor identify patients who have a higher chance to die from prostate cancer and could possibly benefit from more aggressive therapy.
A Canadian multicenter randomized trial designed to compare the effectiveness of two durations of neoadjuvant ADT before EBRT, 353 patients with prostate cancer of any risk group were randomized to receive 3- or 8-month ADT between 1995 and 200121. At a median follow-up of 6.4 years, there was no significant difference in BF, DSS and OS between the study groups. On a post-hoc analysis22, it was suggested that biochemical response measured at the end of ADT is more important than ADT duration before EBRT, when comparing patients whose PSA reached levels ≤0.1 ng/dL versus those who did not reach that level. For the entire cohort, biochemical response to levels ≤0.1 ng/dL was associated with a lower incidence of BF at 8 years; however, it failed to predict DSS or OS in any of the analysis performed for the entire cohort or subgroups; this is likely to be related to the relative short follow-up of this lower-risk prostate cancer population, and the small number of cancer-related deaths in this group of patients. Another possible contributing factor for the lack of predictive power is the threshold value (0.1 ng/dL) chosen for this post-hoc analysis.
Recently D’Amico et al.23 used the Prentice criteria to demonstrate that PSA level higher than 0.5 ng/mL after 6 months of ADT and EBRT, and PSA nadir higher than 0.5 ng/mL, were surrogate endpoints for prostate cancer-specific mortality. This study was performed in a group of patients with intermediate- and high-risk prostate cancer from two different clinical trials: the Dana Farber Cancer Institute, which randomized 206 patients to EBRT alone (70 Gy in 1.8 - 2.0 Gy fractions) or EBRT plus 6 months of ADT (EBRT during months 3 and 4 of ADT); and the Trans-Tasman Radiation Oncology Group Trial 96.01, in which 818 patients were randomized between EBRT alone (66 Gy in 2.0 Gy fractions), EBRT plus 3 months of ADT, or EBRT plus 6 months of ADT (ADT starting 2 months before EBRT in both arms). However, the group receiving 3 months of ADT in the TROG 96.01 trial was not included in the analysis, and a total of 734 patients were assessed: 371 patients treated with EBRT alone and 363 trated with EBRT plus ADT, and were followed for 8.2 years (median). This retrospective analysis shows results that are similar to ours, on a mixed group of patients with prostate cancer and adequate follow-up, but using a smaller number of patients enrolled in two different trials, and a higher PSA cutpoint.
In conclusion, failure to obtain PSA-CR of ≤0.3 ng/mL after short-term hormonal therapy and external beam radiotherapy appears to be an independent predictor of unfavorable outcomes. In an attempt to individualize therapy based on specific tumor needs, this hypothesis generating study suggests that complete responders (i.e., patients with a PSA ≤0.3 ng/dL after short-course ADT and EBRT) could stop ADT earlier than the usually recommended 2–3 years. Such important change in current practice must be investigated on a clinical trial.
Acknowledgments
This trial was conducted by the Radiation Therapy Oncology Group (RTOG), and was supported by RTOG grant U10 CA21661 and CCOP grant U10 CA37422 from the National Cancer Institute (NCI). This manuscript’s contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute.
Footnotes
The authors have no financial disclosure and no conflict of interest related to this work.
References
- 1.Dal Pra A, Cury FL, Souhami L. Combining radiation therapy and androgen deprivation for localized prostate cancer-a critical review. Current Oncology. 2010;17:28–38. doi: 10.3747/co.v17i5.632. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Higano CS. Side effects of androgen deprivation therapy: monitoring and minimizing toxicity. Urology. 2003;61:32–8. doi: 10.1016/s0090-4295(02)02397-x. [DOI] [PubMed] [Google Scholar]
- 3.O’Brien MF, Cronin AM, Fearn PA, et al. Pretreatment prostate-specific antigen (PSA) velocity and doubling time are associated with outcome but neither improves prediction of outcome beyond pretreatment PSA alone in patients treated with radical prostatectomy. Journal of Clinical Oncology. 2009;27:3591–7. doi: 10.1200/JCO.2008.19.9794. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Palma D, Tyldesley S, Blood P, et al. Pretreatment PSA velocity as a predictor of disease outcome following radical radiation therapy. International Journal of Radiation Oncology-Biology-Physics. 2007;67:1425–9. doi: 10.1016/j.ijrobp.2006.11.006. [DOI] [PubMed] [Google Scholar]
- 5.Roach M, 3rd, DeSilvio M, Lawton C, et al. Phase III trial comparing whole-pelvic versus prostate-only radiotherapy and neoadjuvant versus adjuvant combined androgen suppression: Radiation Therapy Oncology Group 9413. Journal of Clinical Oncology. 2003;21:1904–11. doi: 10.1200/JCO.2003.05.004. [DOI] [PubMed] [Google Scholar]
- 6.Roach M., 3rd A phase III trial comparing whole pelvic irradiation followed by a conedown boost to boost irradiation only and comparing neoadjuvant to adjuvant total androgen suppression (TAS) Radiation Therapy Oncology Group. 1998 [Google Scholar]
- 7.Roach M, 3rd, Hanks G, Thames H, Jr, et al. Defining biochemical failure following radiotherapy with or without hormonal therapy in men with clinically localized prostate cancer: recommendations of the RTOG-ASTRO Phoenix Consensus Conference. International Journal of Radiation Oncology-Biology-Physics. 2006;65:965–74. doi: 10.1016/j.ijrobp.2006.04.029. [DOI] [PubMed] [Google Scholar]
- 8.Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Amer Statist Assoc. 1958;53:457–481. [Google Scholar]
- 9.Gaynor JJ, Feuer EJ, Tan CC, et al. On the use of cause-specific and conditional failure probabilities. J Am Stat Assoc. 1993;88:400–409. [Google Scholar]
- 10.Gray RJ. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat. 1988;16:1141–1154. [Google Scholar]
- 11.Cox DR. Regression models and life tables. Journal of the Royal Statistical Society, Series B. 1972;34:187–202. [Google Scholar]
- 12.Roach M, 3rd, 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. Journal of Clinical Oncology. 2008;26:585–91. doi: 10.1200/JCO.2007.13.9881. [DOI] [PubMed] [Google Scholar]
- 13.D’Amico AV, Manola J, 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;292:821–7. doi: 10.1001/jama.292.7.821. [DOI] [PubMed] [Google Scholar]
- 14.Hanks GE, Pajak TF, Porter A, et al. Phase III trial of long-term adjuvant androgen deprivation after neoadjuvant hormonal cytoreduction and radiotherapy in locally advanced carcinoma of the prostate: the Radiation Therapy Oncology Group Protocol 92–02. Journal of Clinical Oncology. 2003;21:3972–8. doi: 10.1200/JCO.2003.11.023. [DOI] [PubMed] [Google Scholar]
- 15.Bolla M, Van Tienhoven G, 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. The Lancet Oncology. 2010;11:1066–73. doi: 10.1016/S1470-2045(10)70223-0. [DOI] [PubMed] [Google Scholar]
- 16.Kuban DA, Levy LB, Cheung MR, et al. Long-term failure patterns and survival in a randomized dose-escalation trial for prostate cancer. Who dies of disease? International Journal of Radiation Oncology-Biology-Physics. 2011;79:1310–7. doi: 10.1016/j.ijrobp.2010.01.006. [DOI] [PubMed] [Google Scholar]
- 17.D’Amico AV, McLeod DG, Carroll PR, et al. Time to an undetectable prostate-specific antigen (PSA) after androgen suppression therapy for postoperative or postradiation PSA recurrence and prostate cancer-specific mortality. Cancer. 2007;109:1290–5. doi: 10.1002/cncr.22550. [DOI] [PubMed] [Google Scholar]
- 18.Ogawa K, Nakamura K, Sasaki T, et al. Postoperative radiotherapy for localized prostate cancer: clinical significance of nadir prostate-specific antigen value within 12 months. Anticancer research. 2009;29:4605–13. [PubMed] [Google Scholar]
- 19.Nickers P, Albert A, Waltregny D, et al. Prognostic value of PSA nadir < or =4 ng/mL within 4 months of high-dose radiotherapy for locally advanced prostate cancer. International Journal of Radiation Oncology-Biology-Physics. 2006;65:73–7. doi: 10.1016/j.ijrobp.2005.11.026. [DOI] [PubMed] [Google Scholar]
- 20.Miller N, Smolkin ME, Bissonette E, et al. Undetectable prostate specific antigen at 6–12 months: a new marker for early success in hormonally treated patients after prostate brachytherapy. Cancer. 2005;103:2499–506. doi: 10.1002/cncr.21077. [DOI] [PubMed] [Google Scholar]
- 21.Crook J, Ludgate C, Malone S, et al. Report of a multicenter Canadian phase III randomized trial of 3 months vs. 8 months neoadjuvant androgen deprivation before standard-dose radiotherapy for clinically localized prostate cancer. International Journal of Radiation Oncology-Biology-Physics. 2004;60:15–23. doi: 10.1016/j.ijrobp.2004.02.022. [DOI] [PubMed] [Google Scholar]
- 22.Alexander A, Crook J, Jones S, et al. Is biochemical response more important than duration of neoadjuvant hormone therapy before radiotherapy for clinically localized prostate cancer? An analysis of the 3- versus 8-month randomized trial. International Journal of Radiation Oncology-Biology-Physics. 2010;76:23–30. doi: 10.1016/j.ijrobp.2009.01.030. [DOI] [PubMed] [Google Scholar]
- 23.D’Amico AV, Chen MH, de Castro M, et al. Surrogate endpoints for prostate cancer-specific mortality after radiotherapy and androgen suppression therapy in men with localised or locally advanced prostate cancer: an analysis of two randomised trials. The Lancet Oncology. 2012;13:189–95. doi: 10.1016/S1470-2045(11)70295-9. [DOI] [PubMed] [Google Scholar]