Delayed Radiotherapy for Patients with Localized Prostate Cancer: Validation by Propensity Score Matching

Patients and Methods

Patients. Patients with localized prostate cancer who underwent radiotherapy and ADT were enrolled in this study. From July 2000 to November 2008, 144 consecutive patients were identified. Patients were referred by six neighbouring hospitals to the Tsukuba Medical Center Hospital, to receive radiotherapy. The median patient age was 71 years, range 52-91 years. The patients' characteristics are shown in Table I. Patients were classified into two groups according to the duration of delay from initial diagnosis by prostate biopsy until the initiation of radiotherapy. Patients for whom radiotherapy was delayed by >6 months were assigned to the delayed group. The other patients were assigned to the non-delayed group. Patients' characteristics between these two groups were well balanced with regard to T-stage, pretreatment prostate specific antigen (PSA) levels, Gleason scores, risk classification, and age.

Pre-treatment evaluations. Pre-treatment evaluations included medical history, performance status, digital examination of the prostate, complete blood count (CBC), hepatic function tests, PSA values, chest X-rays, bone scintigrams, and histological examination of the prostate. Lymph node metastasis was evaluated by abdominal and pelvic computed tomography (CT) and magnetic resonance imaging (MRI). These examinations were performed prior to radiotherapy and ADT. All patients were histologically-diagnosed with adenocarcinoma, with no metastatic abdominal or pelvic lymph nodes, such as the obturator, internal, external, common iliac, or paraaortic lymph nodes. A core biopsy was performed on at least three points in each lobe. T-Stage was diagnosed by MRI and digital rectal examination performed by urologists. Clinical stages were defined based on the 2010 American Joint Committee on Cancer Seventh Edition (5).

After completion of radiotherapy, follow-up was performed every three months for five years, and every 3-6 months thereafter. At each visit, the patients underwent physical examinations and PSA values were assessed.

Radiotherapy. The treatment system included a treatment planning CT (CTS20SPS; Shimadzu Co., Kyoto, Japan), a treatment planning system (RTP; Mitsubishi Electric Co., Tokyo, Japan or Pinnacle3, Philips, Eindhoven, the Netherlands), and a linear accelerator (EXL 20DP; Mitsubishi Electric Co., Tokyo, Japan). All patients received external beam radiotherapy to the whole pelvis, followed by boosted radiotherapy to the prostate using 10-MV photons. Patients were treated five times per week. Between July 2000 and February 2004, the patients were irradiated with up to 46 Gy at 2-Gy daily doses to the whole pelvis, including the prostate, seminal vesicles, and internal iliac lymph nodes, using four-field, three-dimensional, conformal radiotherapy (3DCRT) techniques. Total doses of 70 Gy and 54 Gy were prescribed for the prostate and the seminal vesicles, respectively. After March 2004, the daily dose was reduced to 1.8 Gy, and the total prescribed dose was changed to 46.8 Gy to the whole pelvis, an additional 11.2 Gy to the prostate and the proximal site of the seminal vesicles, and a boosted dose of 14.2-17.6 Gy to the prostate using a seven-field 3DCRT technique. Thus, patients received a total dose of 72.0-75.6 Gy to the prostate. The median dose was 72 Gy given in 40 fractions. The dose-escalating scheme and the number of patients treated are shown in Table II.

Hormone therapy. ADT consisted of a combination of a luteinizing hormone–releasing hormone agonist (LH-RH) and a non-steroidal anti-androgen (flutamide or bicalutamide), or orchiectomy combined with bicalutamide in order to achieve maximal androgen blockade. A long-acting LH-RH was delivered subcutaneously combined with a flutamide dose of 375 mg or a bicalutamide dose of 80 mg orally, daily for six and 119 patients, respectively. Five patients underwent orchiectomy in conjunction with bicalutamide. Four patients received bicalutamide-only. The remaining 10 patients received LH-RH only as hormonal therapy.

End-point definition. The primary end-point for this study was the biochemical progression-free survival. The event of failed progression-free survival was defined as biochemical (PSA) progression. The definition of biochemical progression used in this study was developed by the American Society for Therapeutic Radiology and Oncology in 2005 in Phoenix (6); biochemical progression was defined as an increase of at least 2 ng/ml in nadir PSA level.

Statistical analysis. The χ² test and Fisher's exact test were used to detect differences between the two groups. Student's t-test was used to assess differences in mean values. A p-value of ≤0.05 was considered significant. Biochemical progression-free survival was calculated from the date of diagnosis by prostate biopsy to the date of biochemical progression or last follow-up. Cox proportional hazard regression analysis and univariate analysis were used to identify the most significant independent prognostic factors, using variables such as age, risk classification according to the report of Kuban et al. (7), irradiation dose, and delay in radiotherapy. Survival was calculated by the Kaplan-Meier method, and differences were analyzed with the log-rank test. To validate the results, a propensity score (8) for the delay in radiotherapy was calculated from the coefficients of parameters from a logistic-regression model analysis. The coefficient of parameters included age, irradiated dose, Gleason score, serum PSA levels, and T-stage. After matching patients by their propensity score, biochemical progression-free survival was compared. Statistical analysis between the two groups was performed with STATA 12 (Stata Corp., College Station, TX, USA).