Prostate cancer-specific mortality after definitive radiation therapy: who dies of disease?

BACKGROUND: A competing risks analysis was undertaken to identify subgroups at greatest risk of dying from prostate cancer (PC) after definitive external beam radiation therapy (RT)±androgen deprivation therapy (ADT) in the prostate specific antigen (PSA) era. METHODS: Outcomes of 2675 men with localised PC treated with RT±ADT from 1987-2007 were analysed. Prostate cancer-specific mortality (PCSM) and non-PCSM rates were calculated after stratifying patients according to National Comprehensive Cancer Network (NCCN) risk-group, RT dose, use of ADT and age at treatment. RESULTS: Only 0.2% of low-risk men died of PC 10 years after treatment. All of these deaths occurred in patients treated with < 72 Gy, and only one patient ≥ 70 years old who received ≥ 72 Gy died of PC at last follow-up. Likewise, none of the patients with intermediate-risk disease treated with ≥ 72 Gy and ADT died of PC at 10 years, and the highest 10-year rate of PCSM was seen in men ≥ 70 years old treated with < 72 Gy without ADT (5.1%). Among high-risk men < 70 years old, treatment with RT dose < 72 Gy without ADT yielded similar 10-year rates of PCSM (15.2%) and non-PCSM (18.5%), whereas men treated with ≥ 72 Gy and ADT were twice as likely to die from other causes (16.2%) than PC (9.4%). In high-risk men ≥ 70 years old, dose-escalation with ADT reduced 10-year PCSM from 14% to 4%, and most deaths were due to other causes. CONCLUSION: Low- and intermediate-risk patients treated with definitive RT are unlikely to die of PC. PCSM is higher in men with high-risk disease but may be reduced with dose-escalation and ADT, although patients are still twice as likely to die of other causes.

Sexual function and the use of medical devices or drugs to optimize potency after prostate brachytherapy.

PURPOSE: Prospective evaluation of sexual outcomes after prostate brachytherapy with iodine-125 seeds as monotherapy at a tertiary cancer care center. METHODS AND MATERIALS: Subjects were 129 men with prostate cancer with I-125 seed implants (prescribed dose, 145 Gy) without supplemental hormonal or external beam radiation therapy. Sexual function, potency, and bother were prospectively assessed at baseline and at 1, 4, 8, and 12 months using validated quality-of-life self-assessment surveys. Postimplant dosimetry values, including dose to 10% of the penile bulb (D10), D20, D33, D50, D75, D90, and penile volume receiving 100% of the prescribed dose (V100) were calculated. RESULTS: At baseline, 56% of patients recorded having optimal erections; at 1 year, 62% of patients with baseline erectile function maintained optimal potency, 58% of whom with medically prescribed sexual aids or drugs. Variables associated with pretreatment-to-posttreatment decline in potency were time after implant (p = 0.04) and age (p = 0.01). Decline in urinary function may have been related to decline in potency. At 1 year, 69% of potent patients younger than 70 years maintained optimal potency, whereas 31% of patients older than 70 maintained optimal potency (p = 0.02). Diabetes was related to a decline in potency (p = 0.05), but neither smoking nor hypertension were. For patients with optimal potency at baseline, mean sexual bother scores had declined significantly at 1 year (p < 0.01). Sexual potency, sexual function, and sexual bother scores failed to correlate with any dosimetric variable tested. CONCLUSIONS: Erections firm enough for intercourse can be achieved at 1 year after treatment, but most men will require medical aids to optimize potency. Although younger men were better able to maintain erections firm enough for intercourse than older men, there was no correlation between potency, sexual function, or sexual bother and penile bulb dosimetry.

Improvement in prostate cancer survival over time: a 20-year analysis.

PURPOSE: This study aimed to evaluate the changes in outcome for men with localized prostate cancer treated with definitive external beam radiation therapy during a 20-year period at a comprehensive cancer center. METHODS: We categorized 2675 men with prostate cancer treated at MD Anderson Cancer Center with definitive external beam radiation therapy with or without androgen deprivation therapy into 3 treatment eras: 1987 to 1993 (n = 722), 1994 to 1999 (n = 828), and 2000 to 2007 (n = 1125). To help adjust for stage migration, patients were stratified according to risk group as defined by the National Comprehensive Cancer Network. Biochemical (Phoenix definition), local, distant, and any clinical failure, prostate-cancer specific survival, and overall survival were analyzed according to the Kaplan-Meier method. RESULTS: Median age was 68.5 years and median follow-up was 6.4 years. Fewer men in the most recent era had high-risk disease, and a higher proportion received 72 Gy or higher (99% vs 4%) and androgen deprivation therapy (60% vs 6%) than the earliest era. All risk groups treated in the modern era experienced improved rates of biochemical, local, and distant failure. In high-risk patients, decreased rates of distant failure and clinical failure led to improved prostate cancer-specific survival and overall survival. Local control was improved for intermediate- and high-risk patients, with a trend toward improvement in low-risk patients. On multivariate analysis, recent treatment era was closely correlated with a dose of 72 Gy or higher and treatment with androgen deprivation therapy and predicted for lower rates of biochemical, local, and distant failure. Androgen deprivation therapy, higher dose, and more recent treatment era predicted for improved prostate cancer-specific survival. DISCUSSION: During the last 20 years of prostate cancer irradiation, disease control outcomes have improved in all patients, leading to improved prostate cancer-specific survival and overall survival for men with high-risk disease. This may reflect advances in workup, staging accuracy, and prostate cancer treatment in the modern era.

Clinical and pathologic predictors of locoregional recurrence, distant metastasis, and overall survival in patients treated with chemoradiation and mesorectal excision for rectal cancer.

OBJECTIVES: To identify predictive factors for locoregional recurrence (LR), distant metastasis (DM), and overall survival (OS) in patients treated with chemoradiation and surgery for rectal cancer. METHODS: Between 1989 and 2001, 470 patients with rectal cancer were treated with preoperative (89%) or postoperative (11%) chemoradiation and mesorectal excision. Median radiation dose was 45 Gy; 97% received concurrent infusional 5-fluorouracil, and 65% received adjuvant chemotherapy. Median follow-up interval was 5.7 years. RESULTS: The 5-year rates of freedom from LR, freedom from DM, and OS were 90%, 79%, and 80%, respectively. On univariate analysis, significant predictors of LR were female sex, clinical T stage, pathologic T and N stages, and positive radial margin. Significant univariate predictors of DM were circumferential extent of tumor, tumor immobility, lymphovascular invasion, perineural involvement, and pathologic T and N stages. Significant univariate predictors of lower OS were age, circumferential extent of tumor, shorter distance from anal verge, tumor size, tumor immobility, anal canal involvement, lymphovascular invasion, perineural involvement, positive radial margin, and pathologic T and N stages. On Cox multivariate analysis, female sex and pathologic T and N stages independently predicted for LR; pathologic T and N stages independently predicted for DM; and age, circumferential extent of tumor, positive radial margin, and pathologic T and N stages independently predicted for lower OS. CONCLUSIONS: Pathologic T and N stages significantly predicted for all 3 end points (LR, DM and OS) on multivariate analysis. Investigations of more aggressive adjuvant chemotherapy appear warranted for pathologic stage T3/T4 or N1/2 rectal cancer.

Collecting direct non-health care and time cost data: application to screening and diagnosis of cervical cancer.

BACKGROUND: Data on direct non-health care and time costs are rarely collected, though the incorporation of such data is essential for performing cost-effectiveness analyses according to established guidelines. OBJECTIVES: To explore the challenges involved in collecting and analyzing these data from patients enrolled in a clinical trial. METHODS: Through the use of a pilot study, the authors designed a questionnaire to collect these costs. They used this questionnaire in a clinical trial conducted at a comprehensive cancer center and a public community hospital. Patients in the trial were undergoing screening or diagnostic procedures through a clinical protocol designed to measure the effectiveness of fluorescence and reflectance spectroscopy for detecting cervical precancers. Direct non-health care costs were adjusted to 2003 constant dollars. RESULTS: The authors successfully collected direct non-health care and time cost data, thus demonstrating the feasibility of acquiring such data. Compared to patients receiving diagnostic services for cervical cancer, those receiving screening services for the same condition in both settings incurred lower direct non-health care costs and time costs, as defined in the questionnaire. Compared to patients receiving either service at the comprehensive cancer center, those seeking either service at the public community hospital incurred lower direct non-health care costs and time costs. When outliers were removed, total direct non-health care costs and time costs substantially decreased for diagnostic patients in the comprehensive cancer center; total direct non-health care costs and time costs for other subgroups remained essentially unchanged. CONCLUSIONS: Direct non-health care and time cost data can be collected within a large-scale clinical trial. The setting (community v. specialty hospital) and population (patients receiving screening v. diagnostic examination) makes a difference regarding the cost totals. The order of magnitude of the final result depends on the context in which the non-health care and time cost data will be used.