On the ethical aspects of the testimony of statisticians in court.

This article considers a legal case presented by the Gore campaign to the Florida Eleventh Circuit Court in December 2000 to contest the election results giving the 25 electoral votes from the State of Florida to George W Bush--basically awarding the presidency to Bush. Consideration is given to the ethical aspects of the testimony of two statisticians, one for Gore and one for Bush, who testified about whether the certified election results rejected a number of legal votes and included a number of illegal votes sufficient to place in doubt the results of the election. It is concluded that neither statistician violated accepted ethical standards of the statistical profession, though the Bush statistician's testimony was at least borderline unethical on certain points. The main problem seems to be with the advocacy principle of the American legal system that leads to the prosecution witnesses testifying narrowly to the truths about the case, whereas the defense witnesses do the same from their viewpoint and also try to belittle the testimony of the prosecution witnesses. All of this makes it difficult to uncover the real truths in the case. The judge's decision was to reject Gore's case and this helped to clear the path to the election of Bush as President of the United States.

The role of the biostatistician in cancer research.

This article considers triumphs and challenges for biostatisticians working in oncology at the beginning of the 21st century. The impact of three major articles in biostatistics in the 20th century is considered: Cornfield's 1951 paper on estimating comparative rates from clinical data; Mantel and Haenszel's 1959 paper on obtaining summary measures of relative risk, adjusting for stratification factors in epidemiological studies; and D. R. Cox's 1972 paper, which developed the proportional hazards model for evaluating the effect of covariates on survival time outcomes. Biostatistical challenges for the 21st century are considered for the areas of clinical trials, survival analysis, and statistical genetics.

Biostatistics in the new millennium: a consulting statistician's perspective.

This article attempts to predict the future for biostatistics and biostatisticians in the twenty-first century. Life will certainly be more complex and there will be growth in population, large corporations, and globalization generally. However, there will continue to be problems relating to biology, medicine, health and the environment, so biostatisticians can surely play an important role, if they are willing to adapt to changing circumstances. Personal views are expressed concerning those areas of most recent development that seem likely to be continued in the coming years: applications (clinical trials, epidemiology, vital statistics), philosophies, models, advances in computing, and the profession of biostatistics.

A modification of Simon's optimal design for phase II trials when the criterion is median sample size.

We present a modification to Simon's optimal design for phase II trials in which the objective is to minimize the median sample size rather than the expected sample size when the true response rate is poor (p = p0). We argue that the modified design may be preferred in smaller institutions when the focus is on a single or small number of phase II trials rather than a large program of phase II trials.

Late complications of therapy in 213 children with localized, nonorbital soft-tissue sarcoma of the head and neck: A descriptive report from the Intergroup Rhabdomyosarcoma Studies (IRS)-II and - III. IRS Group of the Children's Cancer Group and the Pediatric Oncology Group.

BACKGROUND: This review of children and adolescents with nonorbital soft-tissue sarcoma of the head and neck was undertaken to describe late sequelae of treatment, as manifested primarily by problems with statural growth, facial and nuchal symmetry, dentition, vision and hearing, and school performance. PROCEDURE: Four hundred sixty-nine patients entered the IRS-II and -III protocols with localized, nonorbital soft-tissue sarcomas of the head and neck from 1978 through 1987. Their overall survival rate was 53% (250/469) at 5 years. Two hundred thirteen patients were surviving relapse-free 5 or more years after diagnosis, for whom there were serial height measurements at 2 or more years after initiation of therapy. Their median age at diagnosis was 5 years; the median length of follow-up was 7 years. All received multiple-agent chemotherapy, and all but 3 received irradiation to the primary tumor volume. Sixty-eight percent of the tumors arose in cranial parameningeal sites, 22% in nonparameningeal sites, and 10% in the neck. We reviewed flow sheets submitted to the IRS Group Statistical Office to ascertain which late sequelae were recorded. RESULTS: One hundred sixty-four patients (77%) had one or more problems recorded. One hundred ninety of the two hundred thirteen patients (89%) were under 15 years of age at study entry, and at follow-up 92 (48%) had failed to maintain their initial height velocity, which had decreased by more than 25 percentile points from the original value. Thirty-six of the one hundred ninety patients (19%) were receiving growth hormone injections. Hypoplasia or asymmetry of tissues in the primary tumor site was reported in 74 patients, and 13 underwent reconstructive surgery. Poor dentition or malformed teeth were noted in 61 patients. Impaired vision developed in 37 patients, owing primarily to cataracts, corneal changes, and optic atrophy. Thirty-six patients had decreased hearing acuity, and 9 were fitted with hearing aids; 5 of these 9 had received cisplatin. Thirty-five patients were noted to have problems learning in school. Four patients developed a second malignancy (two sarcomas, one carcinoma, one leukemia). CONCLUSIONS: Late sequelae affected the majority of these patients treated for soft-tissue sarcoma of the head and neck on IRS-II and -III. The potential impact of certain sequelae could be reduced by specific measures, such as surgical reconstruction and hormonal therapy. Late sequelae must be taken into account in designing future curative treatments.

Immunotherapy with interleukin-2 and alpha-interferon after IL-2-activated hematopoietic stem cell transplantation for breast cancer.

We previously demonstrated findings suggestive of autologous GVHD in patients receiving IL-2-activated peripheral blood stem cells (PBSC) with IL-2 after transplantation. A pilot study was designed to test tolerability, feasibility and frequency of autologous GVHD and engraftment using IL-2 and alpha-IFN post-transplantation. After cyclophosphamide (6 g/m2) and carboplatin (1800 mg/m2), patients with high-risk stage II or III breast cancer received chemotherapy and rhG-CSF mobilized autologous PBSC that had been cultured in IL-2 for 24 h. Subcutaneous administration of IL-2 began on day 0 at 6 x 10(5) IU/m2/day for 5 of 7 days each week and continued for 4 weeks. Once engraftment occurred, alpha-IFN was initiated at a dose of 1 x 10(6)/m2/day subcutaneously for 30 days. Thirty-four consecutive patients with stage II (n=20), IIIA (n=6) and IIIB (n=8) disease were treated. All patients were without evidence of disease at the time of transplantation. The average time required for the ANC to reach 500/mm3 was 10 days (range: 8-11 days) and for platelets to reach 20000/mm3 was 10.7 days (range: 6-21 days). Forty-seven percent of patients (n=16) completed the full course of immunotherapy; the remaining patients received attenuated doses due to patient's request (n=6), development of temperature >38 degrees C (n=3), development of neutropenia (n=3), serious infection (n=1) and miscellaneous reasons (n=5). Four patients experienced transient moderate toxicities (level 3) including elevated liver function tests, nausea, rash and capillary leak syndrome. Pathological findings suggestive of skin GVHD developed in 43% of patients (12/28 patients) when skin biopsies were evaluated in a blinded fashion. At 13 months post-transplant (median; range: 5-24 months), 28 patients (82%) remain disease-free. These results demonstrate the feasibility and toxicity of this regimen along with pathological findings compatible with autologous GVHD of the skin.

Early history of pathology studies by the Intergroup Rhabdomyosarcoma Study Group.

This review chronicles the series of publications that were the result of the willingness of pathologists and clinicians in the United States to share their pathologic materials and clinical data on patients who were placed on treatment protocols for rhabdomyosarcoma and related tumors over an extended period of time. The availability of this database enabled pathologists and clinicians to study a tumor type that is rare in individual institutions, but occurs in large enough numbers to produce valid conclusions not otherwise possible. Furthermore, young investigators were challenged by this opportunity and were able to spend the necessary time to make new observations that, in retrospect, helped direct protocol designs that produced significant improvement in patient survival. The key factor in this process is the surrender of individual scientific prerogatives to a small number of investigators. It is also important to recognize that the pathologist component of these series of contributions is only a part of the entire effort. It takes an organization of gifted, dedicated experts in many disciplines working together. The investigators who served on the Intergroup Rhabdomyosarcoma Group over a 25-year period eminently fulfilled this.

Pretreatment TNM staging of childhood rhabdomyosarcoma: a report of the Intergroup Rhabdomyosarcoma Study Group. Children's Cancer Study Group. Pediatric Oncology Group.

BACKGROUND: The Intergroup Rhabdomyosarcoma Study Group (IRSG) studies began in 1972 and initially used a clinicopathologic system to place patients into prognostic groups. Because of interest in the development of a pretreatment staging system for assessing the posttreatment outcomes of patients with this disease, potential staging elements were retrospectively evaluated in a subset of 505 patients who participated in IRS-II, an IRSG clinical trial. METHODS: Using the IRS-II data, a TNM pretreatment staging system was developed and used to stage prospectively the patients who were entering IRS-III, a subsequent treatment protocol of the IRSG. Failure free survival and overall survival were compared by pretreatment stage in IRS-III as a means of evaluating this TNM staging. RESULTS: The TNM staging system described the tumor (T) in terms of lesion size (< 5 cm or > or = 5 cm) instead of invasiveness, because these two features were not independent of each other. The clinical status of regional lymph nodes (N) was included in the staging system, as was the presence or absence of metastatic disease (M). The latter feature was extremely important, as expected. The anatomic site of the primary tumor also proved to be an important staging element. Classification of patients by tumor size, clinical status of regional lymph nodes, presence or absence of metastatic disease, and location of the primary tumor (at a favorable or unfavorable anatomic site) created four prognostically distinct staging categories that were relatively equal in size. In a prospective evaluation of this staging system with IRS-III patients, the pretreatment staging lost some prognostic impact. The survival of patients with smaller lesions at unfavorable anatomic sites without clinically involved lymph nodes (Stage II) was similar to that of patients with primary tumors at favorable anatomic sites (Stage I). CONCLUSIONS: A pretreatment TNM staging system for childhood rhabdomyosarcoma, developed with data from IRS-II, was not as predictive of patient outcome when applied prospectively to patients treated in the IRS-III trial. These findings could be due to differences in the management strategy used for IRS-III or the statistical variability in the model-fitting process used to develop the staging system. This demonstrates the need for continual reevaluation of staging systems as patient evaluation and treatment innovations are developed.

Radiation therapy for rhabdomyosarcoma: local failure risk for Clinical Group III patients on Intergroup Rhabdomyosarcoma Study II.

PURPOSE: A subset of 362 pediatric patients with rhabdomyosarcoma was selected from a total of 532 eligible IRS-II patients in Clinical Group III to assess the local and regional failure rates following radiotherapy and to determine patient, tumor, and treatment factors contributing to the risk for local and regional failure. METHODS AND MATERIALS: The study population was selected from all eligible IRS-II Clinical Group III patients. Excluded patients were those with "special pelvic" primary sites whose protocol management restricted radiotherapy (n = 123), and those who were removed from the study before radiotherapy was to begin, or because it was omitted (n = 47). A binary recursive partitioning model was used to identify subgroups of the remaining 362 patients at risk of local or regional failure. RESULTS: The local (only) failure rate was 17% (95% confidence interval, 13-21%), and the local (all) failure rate was 20% (95% confidence interval, 16-24%). The 5-year actuarial risk of local (all) failure was 22% (95% confidence interval, 18-27%). The risk of regional (nodal) failure was between 2% and 23%. Increasing tumor size predicted an increased local failure risk. Primary tumors located above the clavicle had a reduced risk of local failure. The binary recursive partitioning model identified a subset of patients at high risk of local failure. Those patients had primary tumors in the chest, pelvic region, extremity, or trunk, or tumors > 10 cm in diameter. Their local failure rate was 35% (compared to 15% for the remaining patients). The subset of patients at high risk for regional (nodal) failure had node involvement at diagnosis and a primary tumor originating at a site other than orbit, parameningeal, or trunk. Compliance with radiation treatment guidelines approached but did not achieve statistical significance as a predictive factor for local failure. By univariate analysis, factors not influencing local failure risk were age, race, gender, adenopathy, and histology. CONCLUSION: Radiation therapy and chemotherapy administered to Clinical Group III patients entered into the IRS-II protocol produced sustained local control in most cases. Knowledge of the factors which predict an increased risk of local or regional failure will facilitate the design of new treatment strategies.

Ewing's sarcoma of soft tissues in childhood: a report from the Intergroup Rhabdomyosarcoma Study, 1972 to 1991.

PURPOSE: One hundred thirty of 2,792 patients (5%) registered on three Intergroup Rhabdomyosarcoma Study clinical trials (IRS-I, -II, and -III) from 1972 to 1991 had an extraosseous Ewing's sarcoma (EOE). We report here the results of multimodality therapy for this tumor. PATIENTS AND METHODS: The 130 patients were less than 21 years of age; 70 (54%) were males. Primary tumor sites were on the trunk in 41 patients, an extremity in 34, the head/neck in 23, the retroperitoneum/pelvis in 21, and other sites in 11. One hundred fourteen patients had no metastases at diagnosis. In 21 patients, the tumor was completely resected; in 30, the localized or regional tumor was grossly resected, and in 63 patients, grossly visible sarcoma was left behind. Sixteen patients (12%) had distant metastases at diagnosis. All patients were given multiagent chemotherapy and most received irradiation (XRT); none were treated with bone marrow transplantation. RESULTS: One hundred seven patients (82%) achieved a complete response. At 10 years, 62%, 61%, and 77% of the patients were alive after treatment on IRS-I, IRS-II, or IRS-III therapeutic protocols, respectively, similar to figures obtained in all IRS patients. At last follow-up evaluation, 42 patients had died of progressive tumor and one of infection. Survival at 10 years was most likely for patients with tumor that arose in the head and neck, extremities, and trunk, and for those who underwent grossly complete tumor removal before initiation of chemotherapy. For patients with localized, gross residual tumor, adding doxorubicin (DOX) to the combination of vincristine, dactinomycin, cyclophosphamide (VAC), and XRT did not significantly improve survival in 39 patients (62% alive at 10 years) compared with that of 24 patients treated with VAC and XRT without DOX (65% alive at 10 years, P = .93). CONCLUSION: This series indicated that EOE in children is similar to rhabdomyosarcoma (RMS) in its response to multimodal treatment. No benefit was apparent from the addition of DOX to VAC chemotherapy in patients with gross residual EOE.

Interleukin-2-activated hematopoietic stem cell transplantation for breast cancer: investigation of dose level with clinical correlates.

Incubating hematopoietic stem cells with IL-2 in vitro for 24 h generates cytotoxic T cells. When infused into patients, these cells may stimulate a graft-versus-tumor (GVT) effect. This clinical trial was designed to assess the ability of IL-2 activated peripheral blood stem cells (PBSC) to reconstitute hematopoiesis, to investigate dose levels and dose-limiting toxicities of IL-2, and to evaluate clinical results and preliminary laboratory effects using a combination of IL-2-activated autologous PBSC followed by IL-2 after transplantation. Sixty-one women with stage II-IV breast cancer were treated. After the administration of carboplatin (200 mg/m2/day for 3 days) and cyclophosphamide (2 g/m2/day for 3 days), patients received autologous PBSC that were cultured in IL-2 for 24 h followed by parenteral administration of IL-2 beginning the day of transplantation. Three escalating doses of IL-2 were evaluated with increasing duration up to 4 weeks. Of the 57 patients receiving IL-2 after tranplantation, 19 patients (33.3%) were unable to complete the planned course of IL-2 therapy due to persistent fevers (n = 9), diarrhea (n = 2), pulmonary capillary leak syndrome (n = 3), development of a rash (n = 1), atrial fibrillation (n = 1), or patient's request (n = 3). One death occurred during hospitalization. Engraftment of neutrophils occurred on day 11.5 (mean; range 8-21 days) and platelets on day 11.7 (mean; range 7-33 days). The maximal tolerated dose of IL-2 was 6 x 10(5) IU/m2/day for 4 weeks. Disease-free survival rates for all stages were comparable to current reports in the literature. Preliminary laboratory evaluations include FACScan analysis of the IL-2 activated PBSC demonstrating an increased percentage of CD3+, CD25+, HLA-DR+ T cells. Phenotypically similar cells were present in peripheral blood samples of patients when tested 15 days after transplantation. This study demonstrates successful engraftment with IL-2-activated PBSC after high-dose chemotherapy for women with stage II-IV breast cancer. The regimen is feasible and, although toxicities are common, they are manageable and correlate with increasing dose and duration of IL-2.

The scientific basis of clinical trials: statistical aspects.

The objective of this paper is to delineate some of the advances in clinical oncology that have occurred since the 1950s in the context of methodological principles established by Dr. Emil J Freireich and colleagues. Four statistical aspects of the methodological developments in clinical trials are considered and illustrated by real examples: a quantitative approach to the design and analysis of clinical trials; the randomized controlled trial; the nonrandomized controlled trial; and the use of regression models in clinical studies.

Classification of rhabdomyosarcomas and related sarcomas. Pathologic aspects and proposal for a new classification--an Intergroup Rhabdomyosarcoma Study.

BACKGROUND: There is a need to develop a single prognostically significant classification of rhabdomyosarcomas (RMS) and other related tumors of children, adolescents, and young adults which would be a current guide for their diagnosis, allow valid comparison of outcomes between protocols carried out anywhere in the world, and should enhance recognition of prognostic subsets. METHOD: Sixteen pathologists from eight pathology groups, representing six countries and several cooperative groups, classified by four histopathologic classification schemes 800 representative tumors of the 999 eligible cases treated on Intergroup Rhabdomyosarcoma Study II. Each tumor was classified according to each of the four systems by each of the pathologists. In addition, two independent subsamples of 200 of the 800 patients were reviewed according to the new system, so that 343 distinct patients were reviewed once, and 57 of these twice. RESULTS: A study of the survival rates of all subtypes in the sample of 800 patients led to the formation of a new system. This was tested on two independent subsets of 200 of the original cases and found to be reproducible and predictive of outcome by univariate analysis. A multivariate analysis of the 343 patients classified according to the new system indicated that a survival model including pathologic classification and known prognostic factors of primary site, clinical group, and tumor size was significantly better at predicting survival than a model with only the known prognostic factors. CONCLUSION: This new classification, termed International Classification of Rhabdomyosarcoma (ICR) by the authors, was reproducible and predictive of outcome among patients with differing histologies treated uniformly on the Intergroup Rhabdomyosarcoma II protocols. We believe it should be utilized by all pathologists and cooperative groups to classify rhabdomyosarcomas in order to provide comparability among and within multi-institutional studies.

Hyperfractionated radiation in children with rhabdomyosarcoma--results of an Intergroup Rhabdomyosarcoma Pilot Study.

PURPOSE: The Intergroup Rhabdomyosarcoma Study (IRS) Group initiated a pilot study (IRS IV-P) of hyperfractionated radiation (HF XRT) with chemotherapy to test the feasibility and toxicity of this combined modality approach in children with localized but nonresected (group III) and metastatic (group IV) rhabdomyosarcoma. METHODS AND MATERIALS: Using the linear quadratic equation, and an alpha/beta ratio of 10 Gy for acute reacting tumor effect and 3 Gy for late reacting normal tissue effect, a HF XRT protocol was developed giving a total radiation dose of 59.4 Gy, in 1.10 Gy fractions, twice daily at 6-8 h intervals. All patients received chemotherapy in addition to irradiation. The radiation scheme was calculated to increase the biologically effective dose to the tumor by 10% without increasing late effects, when compared to a conventional schedule of 50.4 Gy in 1.8 Gy daily fractions. This protocol also was predicted to cause an increase in acute normal tissue effects. RESULTS: Four hundred forty-nine children age 21 years and younger were eligible for the hyperfractionated radiation study of whom 297 had Group III disease and 152 had Group IV disease. A total of 117 patients were excluded from the feasibility and toxicity analysis because of progressive disease or death prior to scheduled irradiation, surgical resection, major protocol violation, treatment with brachytherapy, or missing data. Thus, 332 children were evaluable for the HF XRT protocol. Twenty-eight of the 332 (8%) were given conventional radiation because of physician preference or young age. Twenty of the 332 (6%) were not irradiated because of young age, anesthesia, or transportation problems. All nonirradiated children were < or = 3 years of age. Thus, 284 children, 86% of the evaluable population, received HF XRT. The radiation dose, number of fractions, number of days, and interfractional interval were scored as appropriate in 93% of cases. Review of radiation portals revealed that in 230 of 284 cases (81%) the radiation fields were appropriate, as per protocol. Thus, the HF XRT was feasible treatment in a multiinstitutional study. Analysis of toxicity revealed that 152 of 204 (75%) of Group III and 52 of 80 (65%) of Group IV patients experienced severe or life-threatening toxicity, explained by the addition of chemotherapy with the radiation. The majority of this toxicity was hematopoietic. Observed organ toxicity, which was potentially explained by the radiation treatment, was greatest at the end of radiation, and improved at the 6-week and 3-month evaluation periods. There were no deaths attributed to radiation toxicity and no instance of toxicity that required alteration of the radiation protocol. Thus, the treatment was not associated with toxicity that was considered excessive or unusual. CONCLUSION: The IRS IV-P study confirms that HF XRT combined with chemotherapy is both feasible and tolerable in children with rhabdomyosarcoma. A prospective randomized trial is underway to test its efficacy as compared to conventional radiation among children also receiving concurrent chemotherapy for rhabdomyosarcoma.

The Third Intergroup Rhabdomyosarcoma Study.

PURPOSE: The ultimate goal of the Third Intergroup Rhabdomyosarcoma Study (IRS-III, 1984 to 1991) was to improve treatment outcome in children with rhabdomyosarcoma through clinical trials comparing risk-based protocols of surgery and multiagent chemotherapy, with or without irradiation. PATIENTS AND METHODS: One thousand sixty-two previously untreated, eligible patients who were entered onto the study after surgery were randomized or assigned to treatment by clinical group (I through IV), histology (unfavorable or favorable), and site of the primary tumor. Initial responses, progression-free survival (PFS), and survival (S) were the end points used in comparisons between randomized groups and between patients treated in IRS-III and IRS-II (1978 to 1984). RESULTS: The overall outcome of therapy in IRS-III was significantly better than in IRS-II (5-year PFS, 65% +/- 2% v 55% +/- 2%; P < .001 by stratified testing). Patients with group I favorable-histology tumors fared as well on a 1-year regimen of vincristine and dactinomycin (VA), as did a comparable group treated with VA plus cyclophosphamide (C) (5-year PFS, 83% +/- 3% v 76% +/- 4%; P = .18). Results for patients with group II favorable-histology tumors, excluding orbit, head, and paratesticular sites, were inconclusive regarding the benefit from addition of doxorubicin (ADR) to VA. Patients with group III tumors, excluding those in special pelvic, orbit, and other selected nonparameningeal head sites, fared much better on the more intensive regimens of IRS-III than on pulsed VAC or VAC-VADRC in IRS-II (5-year PFS estimates, 62% +/- 3% v 52% +/- 3%; P < .01); however, there were no significant differences in outcome among the groups treated in IRS-III. Patients with metastatic disease at diagnosis (clinical group IV) did not benefit significantly from the more complex therapies evaluated in IRS-III. CONCLUSION: Intensification of therapy for most patients in IRS-III, using a risk-based study design, significantly improved treatment outcome overall. The largest gain from this strategy was realized in patients with gross residual tumor after biopsy (clinical group III). It was also possible to decrease therapy for selected patient subsets without compromising survival.

Agreement among and within groups of pathologists in the classification of rhabdomyosarcoma and related childhood sarcomas. Report of an international study of four pathology classifications.

BACKGROUND: An International Pathology study was conducted to measure the agreement demonstrated among and within groups of pathologists involved in the categorization of childhood rhabdomyosarcoma according to four pathology classifications. Data concerning agreement and survival experience according to patho-new subtypes were used as a basis for selection of a proposed new pathologic classification. METHODS: A random sample of 800 eligible patients was chosen from the Intergroup Rhabdomyosarcoma Study II (IRS-II) and was reviewed by pathologists representing eight institutions. A 20% sample of the 800 patients was then reviewed by the pathologists to determine the level of agreement with their original classification. In each instance the patients were classified according to four pathology systems: the conventional system, the International Society for Pediatric Oncology system (SIOP), the National Cancer Institute (NCI) system, and the cytohistologic system. RESULTS: Among the groups of pathologists, the highest measure of agreement was a Kappa value of K = 0.451 for the conventional system, followed by K = 0.406 for the SIOP system, K = 0.384 for the NCI system, and K = 0.328 for the cytohistologic system. For reproducibility within the groups of pathologists, the highest measure of agreement was K = 0.605 for the conventional system, followed by K = 0.579 for the NCI system, K = 0.573 for the SIOP system, and K = 0.508 for the cytohistologic system. CONCLUSIONS: There was a general similarity between the agreement reached within the modified conventional, STOP, and NCI systems, with the modified conventional system having the highest Kappa values, and thus the highest measure of agreement, both among and within the groups of pathologists. Also, the subtypes of the conventional system demonstrated a highly significant relationship to survival time. Hence, based on criteria of reproducibilty and prognostic significance, the proposed classification will essentially be a modification of the conventional system with elements of the SIOP and NCI systems.

An optimal three-stage design for phase II clinical trials.

A phase II clinical trial in cancer therapeutics is usually a single-arm study to determine whether an experimental treatment (E) holds sufficient promise to warrant further testing. When the criterion of treatment efficacy is a binary endpoint (response/no response) with probability of response p, we propose a three-stage optimal design for testing H0: p < or = p0 versus H1: p > or = p1, where p1 and p0 are response rates such that E does or does not merit further testing at given levels of statistical significance (alpha) and power (1--beta). The proposed design is essentially a combination of earlier proposals by Gehan and Simon. The design stops with rejection of H1 at stage 1 when there is an initial moderately long run of consecutive treatment failures; otherwise there is continuation to stage 2 and (possibly) stage 3 which have decision rules analogous to those in stages 1 and 2 of Simon's design. Thus, rejection of H1 is possible at any stage, but acceptance only at the final stage. The design is optimal in the sense that expected sample size is minimized when p = p0, subject to the practical constraint that the minimum stage 1 sample size is at least 5. The proposed design has greatest utility when the true response rate of E is small, it is desirable to stop early if there is a moderately long run of early treatment failures, and it is practical to implement a three-stage design. Compared to Simon's optimal two-stage design, the optimal three-stage design has the following features: stage 1 is the same size or smaller and has the possibility of stopping earlier when 0 successes are observed; the expected sample size under the null hypothesis is smaller; stages 1 and 2 generally have more patients than stage 1 of the two-stage design, but a higher probability of early termination under H0; and the total sample size and criteria for rejection of H1 at stage 3 are similar to the corresponding values at the end of stage 2 in the two-stage optimal design.

The Intergroup Rhabdomyosarcoma Study-II.

BACKGROUND: Intergroup Rhabdomyosarcoma Study (IRS)-II, (1978-1984) had the general goals of improving the survival and treatment of children with rhabdomyosarcoma (RMS). METHODS: Nine hundred ninety-nine previously untreated eligible patients entered the study after surgery and were randomized or assigned to therapy by IRS Clinical Group (I-IV), tumor site, and histologic type. Outcomes were compared between treatments and with results of IRS-I (1972-1978). RESULTS: Patients in Group I, excluding extremity alveolar (EA) RMS, were randomized to standard vincristine (V), dactinomycin (A), and cyclophosphamide (C) or standard VA. At 5 years, disease-free survival (DFS) and survival (S) rates were similar between VAC and VA (DFS: 80%, 70%, P = 0.47; S: 85%, 84%, P = 0.73). Patients in Group II, excluding EA RMS, received radiation and were randomized to intensive VA or repetitive-pulse VAC. Outcomes were similar for rates of DFS (69%, 74%, P = 0.83) and S (88%, 79%, P = 0.17). Patients in Group III, excluding certain pelvic tumors, received radiation and were randomized to repetitive-pulse VAC or repetitive-pulse VAdrC-VAC (Adr, Adriamycin [doxorubicin]). Complete remission (CR) rates were close at 74%, 78%, respectively (P = 0.32), as were percentages in CR (73%) and S (66%) rates; the latter outcomes were significantly better than IRS-I (CR: 56%, P < 0.001; S: 50%, P < 0.001). Central nervous system prophylaxis for Group III patients with cranial parameningeal sarcoma increased S rate to 67% from 45% in IRS-I (P < 0.001). Patients in Group IV received the same regimens as Group III; the CR rate was 53%, 38% remained in CR and S rate was 27% with and 26% without Adr (P = 0.90). At 5 years, S rate for IRS-II, including EA and all pelvic tumors, was 63%: an 8% increase over IRS-I (P < 0.001). Outcomes by primary site were as good as, or better than, the IRS-I experience. CONCLUSIONS: Combining all Groups and treatments in IRS-II, the major improvement in S rate at 5 years between studies was in nonmetastatic patients (71% for IRS-II versus 63% for IRS-I, P = 0.01).

Plain radiographic predictors of survival in treated Ewing's sarcoma. IESS Committee.

We analyzed 16 radiographic features of primary Ewing's sarcoma in 342 patients from the IESS 7299 (Intergroup Ewing's Sarcoma Study) for prognostic significance. Of these, 3 features demonstrated a statistically significant relationship to survival: maximal tumor dimension, tumor location, and an appearance of honey-combing within the lesion. Those individuals with primary lesions centered in the pelvis, femur, or humerus had a worse survival rate than those with lesions centered elsewhere. Similarly, we observed an inverse relationship between the greatest dimension of the osseous portion of the lesion and survival. The radiographic finding of honeycombing also showed a statistical association with improved survival. However, this feature was observed infrequently, and its significance should be assessed further. Otherwise, no other feature that we analyzed showed any relationship to patient prognosis.