A guinea pig''s view on prostate cancer screening trials

Although the combination of paclitaxel with doxorubicin has yielded high response rates in metastatic breast cancer, severe cardiotoxic events have been reported in several patients. The rationale for our study was to evaluate the activity of paclitaxel/doxorubicin combination in patients with this disease but to avoid excessive cardiotoxicity. Therefore, we administered 4 cycles of doxorubicin/paclitaxel followed by 6 cycles of standard cyclophosphamide, methotrexate and 5-fluorouracil (CMF) regimen. Study medication consisted of doxorubicin 60 mg/m2 as a 15-min intravenous infusion followed by paclitaxel 175 mg/m2 as a 3-hour infusion. CMF regimen consisted of cyclophosphamide 600 mg/m2 as 1-hour intravenous infusion followed by methotrexate 40 mg/m2 and 5-fluorouracil 600 mg/m2 bolus injection. The main toxicity of doxorubicin/paclitaxel treatment phase was neutropenia (WHO grade 3/4, 58%), but we observed only one cardiac adverse event. Toxicities of the CMF treatment phase were not significant. Of 24 patients evaluable for response, 2 (8%) had complete responses and 11 (46%) achieved partial response. Ten additional patients (42%) had stable disease. The median time to progression was 12 months and the median overall survival was 18.5 months. The sequential administration of doxorubicin and paclitaxel followed by CMF appeared active and well tolerated in patients with metastatic breast cancer.     

Meta-analysis of clinical trials for chemotherapy in cancer

Meta-analysis has attracted great interest among clinical practitioners in recent years, leading to a steady output of related publications. Meta-analytic articles are easily found in the MEDLINE database using the publication-type option. This paper reviews how to use and understand meta-analysis with a special reference to chemotherapy applied to cancer patients. It is described in relationship to evidence-based medicine (EBM) and clinical practice guidelines. Cochrane collaboration is also referred to as an active voluntary organization conducting meta-analysis. In the technical sections, statistical issues and graphic representations are clearly illustrated using the example of hepatic arterial infusion for colorectal cancer patients. The difference between fixed and random effects models is briefly explained. Finally, an example from Cochrane Library, namely progestagen therapy for endometrial cancer, is illustrated to show the implications of meta-analysis for clinical practice.     

Prostate cancer clinical trials of the Southwest Oncology Group

The changing clinical dynamics of prostate cancer have resulted in a broadening of the research focus of the Genitourinary (GU) Cancer Committee of the Southwest Oncology Group (SWOG). Beginning with an emphasis on hormone-refractory disease in its early years, SWOG prostate cancer trials now cover the entire spectrum of the disease: localized, locally advanced, metastatic and hormone-refractory disease. As the world's largest GU cancer research group, the GU committee of SWOG has pioneered studies in combined androgen therapy for metastatic disease, quality-of-life (QOL) assessments for patients with localized and advanced disease, adjuvant therapy models, and prostate cancer chemoprevention. The committee has also formed the GU Global Group, whose purpose is to convene the chairs of the GU committees of all the major national and international oncology cooperative groups. Meeting semiannually, this group discusses activities within their respective organizations, plans collaborative strategies and protocols, and establishes global strategy in prostate cancer clinical research. The future directions of national and international prostate cancer trials will build on this broad foundation of well-conceived, logically sequenced studies.     

Bayesian interim analysis of phase II cancer clinical trials

Many popular sequential phase II clinical trial designs optimize some criterion subject to constraints on the error probabilities at null and alternative values of the response rate. Such designs may forfeit optimality if one fails to conduct analyses strictly according to plan. Moreover, a decision, say, to accept the experimental therapy at one interim analysis does not necessarily imply the same degree of evidence as the same decision when made at another analysis. I propose an alternative design that bases decisions on the ability of the data to persuade either a sceptic or an enthusiast. My standard of evidence, called the persuasion probability, is based on the Bayesian posterior probability that the experimental treatment is superior to the standard. The design calls for termination at any interim analysis at which an observed persuasion probability exceeds its critical value. I investigate the standards of evidence implied by some frequentist procedures and calculate frequentist properties of persuasion-probability designs.     

Patient attitudes regarding treatment-related erectile dysfunction at time of early detection of prostate cancer

OBJECTIVES: To assess potency rate and patient attitudes regarding erectile dysfunction. METHODS: A multiple choice, self-administered questionnaire distributed to 750 men undergoing testing for early detection of prostate cancer was used. RESULTS: Overall, 33.9% of patients reported either partial or complete lack of erections and 31.1% were not sexually active or active less than once per month. Furthermore, 55.4% would be affected or very affected by lack of erections and 73.6% chose definitive treatment despite a 50% chance of erectile dysfunction. Finally, 47.4% found such treatment-induced erectile dysfunction to be an important or very important problem. When asked to ascribe a quantity of life or period of time that they would be willing to sacrifice to preserve sexual function following treatment, only 15.2% of patients were able to do so, but no consensus could be reached regarding its value. CONCLUSIONS: Reported differences in quality-adjusted life expectancy when screening was compared to no screening and definitive therapy was compared to expectant management are marginal. Therefore, close attention to seemingly minor variables such as existing impotence rate, attitude regarding erectile dysfunction, and willingness to undergo therapy despite its inherent morbidity may substantially reduce or even reverse this reported disadvantage.     

Apples and oranges: building a consensus for standardized eligibility criteria and end points in prostate cancer clinical trials

PURPOSE: To survey eligibility and response criteria for clinical trials in hormone-refractory prostate cancer (HRPC). METHODS: Thirty-five established investigators of HRPC completed a 125-question survey. RESULTS: There was a general consensus that criteria for clinical trial entry would include progression based on an increasing prostate-specific antigen (PSA) level (94% of investigators), an increase in measurable disease (91%), and/or appearance of new bone lesions on bone scan (83%). Most believed that castrate levels of testosterone (77%) and progression after antiandrogen withdrawal (97%) should be documented before study enrollment. Continuation of testicular androgen suppression would be required by 82%. Seventy-seven percent favored separate reports on response rates in bone, measurable disease, symptoms, and biochemical markers (primarily PSA levels), rather than a composite response. Ninety-four percent of the investigators accepted changes in PSA level as a surrogate end point of response. However, interpretation by these investigators of a PSA data set similar to what might be observed in a clinical trial showed marked discordance. Survival is the end point of most importance to 94% of these investigators. Response based on changes in measurable disease, time to progression, response duration, PSA level decrease, or quality-of-life improvement were of similar weighted value as a clinical trial end point and were rated as less important to these investigators than survival (P < 10(-8)). CONCLUSION: This survey indicates some consensus on eligibility and concomitant treatments for clinical studies in HRPC. The use of multiparameter assessment of response and PSA level as a surrogate end point have been widely adopted.     

Attitudes of patients to randomised clinical trials of cancer therapy

The aim of this study was to test an instrument which might be useful for doctors in explaining the randomisation procedure to an individual patient. The sample comprised 323 patients with cancer attending for out-patient appointments and/or chemotherapy treatment in two major cancer centres in the U.K. 315 patients completed a self-report questionnaire--The Attitudes to Randomised Trials Questionnaire (ARTQ). The results show that the majority of subjects 287 (91.1%) believe that patients should be asked to take part in medical research, but only 242 (76.8%) would be prepared to take part in a study comparing two treatments. If treatment was randomised, only 141 (44.8%) would agree to participate. When given further information about the randomisation procedure, 119 (68.4%) of the 174 (55.2%) who initially said 'no' to randomisation or who were unsure, would change their minds and take part in a trial. The ARTQ discriminated between three categories of patient with the following prevailing attitudes: (a) those who seem comfortable with the concept of randomisation; (b) those with some concerns, who with fuller explanation are prepared to consider randomisation; and (c) those firmly against randomisation and participation in trials whatever information is provided. Prior knowledge of patients' attitudes might assist communication about trials and encourage more doctors to approach eligible patients.     

Epidemiology and clinical features of prostate cancer in Japan

The incidence rate of clinically manifest prostate cancer in 1992 was estimated 15.7 per 100,000 men, although it is increasing exponentially. Accordingly, 5399 deaths from prostate cancer in 1995 will be increased to 13,494 deaths in 2015. Change in dietary habit (more Western-style diet) is considered to be a major cause of the increase. Escalating number of elderly people in the Japanese population is another major reason of elevated incidence. On the other, public awareness of prostate cancer and introduction of serum PSA measurement to health check-up undoubtedly have raised the detection rate of early stage disease. The way of androgen ablation do not seem to have influenced on survival of the advanced disease so far. It remains to be clarified whether the combined androgen blockade using pure anti-androgens with castration provide better patients' survival than castration alone.     

Problems associated with randomized controlled clinical trials in breast cancer

The domination of the randomized controlled clinical trial in clinical research in breast cancer is questioned in the context of adjuvant therapy. The criteria for selection of cases by TNM classification are shown to be poorly related to tumour behaviour. The factors that determine outcome--the presence or absence of metastases, their size at the time of presentation of the primary cancer and their growth rate--are not measured. Metastases cannot be detected at sizes below about 10 mm in diameter. This represents about 30 doublings of volume or three-quarters of the life span of the tumour. There is no identifiable starting point for a trial. It is possible to estimate growth rates from the rates of shrinkage in response to primary medical treatment. Modelling of the times at which metastases may appear in the clinic from different possible starting sizes reveals that the usual end points of time from primary diagnosis to metastasis or death have an enormous intrinsic variability. Treatment is merely another confounding variable. The randomized controlled clinical trial is shown to be a poor tool for the investigation of adjuvant treatments. An alternative approach, based upon observation of individual tumour response to primary medical treatment, is commended.     

A similar pattern of chromosomal alterations in prostate cancers from African-Americans and Caucasian Americans

A combination of genetic and epigenetic factors may explain the disproportionate incidence and mortality of prostate cancer among African-American males (AAMs) as compared with Caucasian American males (CAMs). We wished to determine whether primary prostate cancers from AAMs and CAMs harbor different patterns or frequencies of chromosomal alterations. Comparative genomic hybridization (CGH) was performed on clinically localized, untreated primary prostate cancers from 16 AAMs and 16 CAMs. Detailed statistical analysis was used to delineate gains and deletions with high sensitivity and specificity and to compare the frequency and pattern of alterations between the two groups of tumors. The two groups of patients had indistinguishable preoperative serum prostate-specific antigen levels, and the two groups of tumors had similar pathological stages and grades. Chromosomal gains and deletions occurred in regions known to be frequently altered in prostate cancer. Specifically, the most frequent alterations were deletions of regions on chromosomes 13q, 5q, 16q, and 8p and gains of regions on 8q and 5q. When tumors from AAMs and CAMs were compared, the frequencies of alteration (deletion, gain, or no alteration) were similar across 98.9% of the length of the genome. The patterns of alterations of the most frequently altered chromosomes were also similar between tumors from AAMs and CAMs. We concluded that primary prostate cancers from AAMs and CAMs harbor a similar pattern and frequency of chromosomal alterations. These data support the notion that sporadic prostate cancers from AAMs and CAMs develop by similar chromosomal mechanisms. Biological differences, if present, do not occur on the chromosomal level.     

Self-designing clinical trials

I present a method of sequential analysis for randomized clinical trials that allows use of all prior data in a trial to determine the use and weighting of subsequent observations. One continues to assign subjects until one has 'used up' all the variance of the test statistic. There are many strategies to determine the weights including Bayesian methods (though the proposal is a frequentist design). I explore further the self-designing aspect of the randomized trial to note that in some cases it makes good sense (i) to change the weighting on components of a multivariate endpoint, (ii) to add or drop treatment arms (especially in a parallel group dose ranging/efficacy/safety trial), (iii) to select sites to use as the trial goes on, (iv) to change the test statistic and (v) even to rethink the whole drug development paradigm to shorten drug development time while keeping current standards for the level of evidence necessary for approval.