Analysis of the probability and risk of cause-specific failure.

PURPOSE: Kaplan-Meier curves are frequently misused in the analysis of nonsurvival endpoints, such as time to local failure or time to late complications. More appropriate analyses are available and described. METHODS AND MATERIALS: Cumulative incidence is an unbiased estimate of probability of cause-specific failure. Cumulative conditional probability of cause-specific failure reflects risk to patients remaining at risk. Hazard rates also measure risk. RESULTS: Kaplan-Meier curves overestimate the probability of late complications when there is a high mortality rate. Cumulative incidence and cumulative conditional probability accurately give the probability and risk of cause-specific failure. CONCLUSION: Kaplan-Meier analysis of cause-specific failure should be avoided because of its misinterpretation as an estimate of probability, in favor of appropriate methods.

Fast-neutron therapy in advanced head and neck cancer: a collaborative international randomized trial.

PURPOSE: To compare the efficacy of fast-neutron radiotherapy with that of conventionally fractionated photon therapy in the management of patients with locally advanced squamous cell carcinoma of the head and neck. METHODS AND MATERIALS: Patients with Stage III or IV disease were randomized to receive either 20.4 Gy/12 fractions/4 weeks of neutrons or 70 Gy/35 fractions/7 weeks of photons (control). Between April 1986 and March 1991, 178 patients were entered, 169 of whom were eligible for analysis. The treatment arms were balanced for age, stage, and performance status, but not for primary site of origin. RESULTS: Complete response occurred in 70 and 52% with neutrons and photons, respectively (p = 0.006). Local regional failure at 3 years for all patients was 63% for neutrons and 68% for photons. Actuarial overall survival curves were virtually identical in both study arms, falling to 27% at 3 years. Acute toxicity was similar in the two arms, but late grade 3-5 toxicity was 40% with neutrons compared to 18% with photons (p = 0.008). CONCLUSION: Although the initial response rate was higher with neutrons, permanent local control and survival were not improved, and the incidence of late normal tissue toxicity was increased. As a result, fast-neutron therapy for advanced squamous cell carcinoma of the head and neck can only be recommended for patients in whom the logistic benefit of treatment in 12 sessions over 4 weeks outweighs the increased risk of late toxicity.

Photon versus fast neutron external beam radiotherapy in the treatment of locally advanced prostate cancer: results of a randomized prospective trial.

PURPOSE: To evaluate the effectiveness of fast neutron radiation therapy in treatment of locally advanced carcinomas of the prostate. METHODS AND MATERIALS: From April 1986 to October 1990, 178 patients were entered on a prospective, multi-institutional randomized study of the NCI-sponsored Neutron Therapy Collaborative Working Group. This trial compared external beam photon irradiation (7000-7020 cGy) with external beam neutron irradiation (2040 ncGy) for patients with high-grade T2 or T3-4, N0-1, M0 adenocarcinomas of the prostate. Eighty-nine patients were randomized to each treatment. Six patients were subsequently judged to be ineligible, leaving 85 photon and 87 neutron randomized patients eligible for analysis. RESULTS: With a follow-up time ranging from 40 to 86 months (68 months median follow-up) the 5-year actuarial clinical local-regional failure rate for patients treated with neutrons was 11%, vs. 32% for photons (p < 0.01). Incorporating the results of routine posttreatment prostate biopsies, the resulting "histological" local-regional tumor failure rates were 13% for neutrons vs. 32% for photons (p = 0.01). To date, actuarial survival and cause-specific survival rates are statistically indistinguishable for the two patient cohorts, with 32% of the neutron-treated patient deaths and 41% of the photon-treated patient deaths caused by prostate cancer (p = n.s.). Prostate specific antigen (PSA) values were elevated in 17% of neutron-treated patients and 45% of photon-treated patients at 5 years (p < 0.001). Severe late complications of treatment were higher for the neutron-treated patients (11% vs. 3%), and were inversely correlated with the degree of neutron beam shaping available at the participating institutions. Neutron treatment delivery utilizing a fully rotational gantry and multileaf collimator did not result in an increase in severe late effects when compared to photon treatment. CONCLUSION: High energy fast neutron radiotherapy is safe and effective when adequate beam delivery systems and collimation are available, and it is significantly superior to external beam photon radiotherapy in the local-regional treatment of large prostate tumors.

Zoladex (goserelin acetate implant) in the treatment of endometriosis: a randomized comparison with danazol. The Zoladex Endometriosis Study Group.

OBJECTIVE: To compare the efficacy, endocrine effects, and safety of Zoladex (goserelin acetate) and danazol in the treatment of premenopausal women with endometriosis in a multicenter, randomized, open study. METHODS: Three hundred fifteen patients with stages I-IV endometriosis (revised American Fertility Society [AFS] classification) were treated with Zoladex, 3.6 mg every 28 days by subcutaneous injection, or danazol, 400 mg orally twice daily for 24 weeks. Efficacy was assessed by determination of pelvic signs and symptoms scores and revised AFS endometriosis scores. Endocrine effects were determined by measurements of hormone levels. Safety was evaluated by physical examination, laboratory indices, occurrence of adverse events, and bone mineral density changes. RESULTS: Both treatments significantly (P < .0001) reduced mean subjective signs and symptoms scores both during and after therapy. The mean percent reduction in the revised AFS endometriosis score after 24 weeks of treatment was 53% for Zoladex and 33% for danazol, and reduction in the endometrial implants score was 56% for Zoladex and 46% for danazol. Serum estradiol levels decreased to the postmenopausal range in the Zoladex group and to the early follicular phase range in the danazol group. Hypoestrogenic effects occurred more frequently with Zoladex, whereas androgenic side effects were more common with danazol. There was a higher percentage of withdrawals due to adverse events with danazol than with Zoladex. Mean bone mineral density decreased from baseline by 5.4% in the Zoladex group and increased by 1.0% in the danazol group at the end of treatment. CONCLUSION: Zoladex is as well tolerated and as effective as danazol in the treatment of premenopausal women with endometriosis.

Goserelin versus orchiectomy in the treatment of advanced prostate cancer: final results of a randomized trial. Zoladex Prostate Study Group.

OBJECTIVES: To compare the efficacy and safety of goserelin and orchiectomy in patients with stage D2 prostate cancer. METHODS: A randomized, open, multicenter study was conducted in 283 patients. Patients were allocated to goserelin, 3.6 mg every 28 days or to orchiectomy. Study end points were endocrine response, objective response, time to treatment failure, survival, and tolerability. Objective response was based on modified criteria of the National Prostate Cancer Project. RESULTS: Serum testosterone decreased from baseline to castrate levels by week 4 in each group and remained below castrate levels thereafter. Acid phosphatase and alkaline phosphatase concentrations also decreased in each group. The goserelin and orchiectomy groups had similar results for objective response (82% versus 77%) and had similar medial times to treatment failure (52 versus 53 weeks) and survival (119 versus 136 weeks). No significant interactions between treatments and prognostic factors were observed. Adjusting for baseline testosterone concentration had no effect on survival outcome. Race had no influence on outcome or efficacy end points. Common adverse events in both groups were pain, hot flushes, and lower urinary tract symptoms. CONCLUSIONS: Goserelin is well tolerated and as effective as orchiectomy in patients with Stage D2 prostate cancer.

Increased mortality rates after cataract surgery. A statistical analysis.

One hundred ninety-three patients 50 to 89 years of age had cataract surgery between January 1, 1979, and December 31, 1980, at West Virginia University Medical Center, Morgantown. In comparison with 182 patients who elected one of three other surgical procedures, cataract surgery patients had a significantly higher mortality rate (P = 0.0005) than control group patients, according to life-table analysis estimates adjusted for age and sex. Patients with adult onset diabetes mellitus had slight increased survival and did not alter relative mortality. The authors' results support the hypothesis that senile cataracts may reflect systemic factors in addition to localized ocular disease.

A generalized effective exposure modeling program for assessing dose-response in epidemiologic investigations.

The computer program OCMAP (occupational cohort mortality analysis program) has been extended to allow exposure indices for person-years to be calculated as time-dependent covariates rather than as the usual cumulative indices. The article develops a general formula for weighting exposures and describes the cumulative index, the lag, partial weight, and the lognormal models as specific examples. The implementation in OCMAP is flexible in that it permits these specific models or any other user-specified weighting scheme to be used. Several example schemes are applied to a cohort of workers exposed to arsenic.

Independent prognostic factors in patients with metastatic (stage D2) prostate cancer. The Zoladex Study Group.

The independent prognostic factors affecting survival were assessed in 240 men undergoing treatment for metastatic prostate cancer as part of a randomized clinical trial comparing the gonadotropin releasing hormone analogue Zoladex (goserelin acetate implant) with castration. In a multivariate analysis, the most highly significant predictors were the presence or absence of bone pain, serum testosterone levels, serum alkaline phosphatase levels, and performance status. Patients with all four factors favorable for survival had a 2-year survival rate of 84% as compared with only 8% for patients with none of the four factors favorable for survival. No other factors were significant. A separate analysis of serum testosterone levels revealed that the higher the pretreatment serum testosterone level, the greater the survival rate. Compared with patients with serum testosterone levels less than 6.9 nmol/L, significant differences in survival were observed for patients with serum testosterone levels of 10.4 to 13.9, 13.9 to 17.3, and over 17.3 nmol/L. These results have important implications for the design and analysis of future clinical trials of hormone therapy and for counseling patients regarding the short-term prognosis of their disease.

Leukocyte interferon for treating first episodes of genital herpes in women.

Women experiencing their first episodes of genital herpes were treated, beginning within three days of the onset of lesions, with 5 X 10(4) units of human leukocyte interferon/kg of body weight for 12 doses over 14 days (total, approximately 3.6 X 10(7) units) or with placebo in equivalent volumes. Life-table analysis revealed quicker healing and significant reductions in the duration of shedding of virus in interferon-treated patients. Maximum daily geometric mean titers of virus and total area of unhealed lesions also decreased more quickly. No statistically significant difference in resolution of pain was seen between the two groups. Interferon had no effect on onset or frequency of subsequent recurrences recorded over one year of follow-up. Moderate, transient neutropenia occurred in 13 of 34 interferon-treated patients. A therapeutic effect of human leukocyte interferon on initial genital herpes was documented, but the clinical usefulness of interferon treatment of genital herpes is limited at this time.

Treatment related sequelae following external beam radiation for prostate cancer: a review with an update in patients with stages T1 and T2 tumor.

The primary goal of radical radiation therapy in men with localized prostate carcinoma is cure and a secondary but important goal is to achieve cure without treatment related sequelae, such as loss of continence, rectal injury, loss of potency and the need for castration. A literature review of 2,611 men undergoing irradiation for all stages of localized prostatic carcinoma documented a 0.2% incidence of treatment related mortality, 1.9% severe complications, 0.9% incontinence and 33 to 60% maintenance of full potency 5 or more years after treatment. A separate analysis was made of 331 patients with only early tumors (stages T1 and T2) treated with conventional external beam radiation therapy to doses of 63 to 74 Gy. from 2 individual centers (Massachusetts General Hospital and M.D. Anderson Hospital) and 1 multi-institutional group (Radiation Therapy Oncology Group). Median followup was 6.1 years; however, in 2 series followup ranged to 14 years. This analysis revealed frequencies of treatment associated sequelae of 0% for mortality, 0% severe complications, 0.4% urinary incontinence, 5.4% genitourinary structures (1.2% persisting), 5.1% hematuria (0.9% persisting) and 5.4% rectal bleeding (0.6% persisting). This composite analysis of men undergoing irradiation for stages T1 and T2 tumors with conventional fractionation and doses indicates that acute morbidity is minor and usually transient, severe injury is rare, most late gastrointestinal and genitourinary symptoms of radiation injury are neither permanent nor debilitating, and few symptoms of radiation injury develop beyond 5 years from treatment. These data, combined with the low progression rates (using prostate specific antigen criteria) following irradiation of men with early tumors, further substantiate the primary role of radical radiotherapy in the treatment of surgical risk adversive patients.