Interim monitoring for non-inferiority trials: minimizing patient exposure to inferior therapies.

The goal of a non-inferiority randomized trial is to demonstrate that an experimental treatment is not unacceptably worse than a standard treatment. The experimental treatment is known to have less toxicity or other quality-of-life benefits when compared with the standard treatment, so that a small decrement in efficacy would be acceptable. Interim monitoring of randomized trials is used to stop trials early if the conclusions of the trial become definitive early. In the context of a non-inferiority trial, of special interest is stopping a trial early when the experimental treatment is inferior to the standard treatment. Methods for performing interim monitoring of non-inferiority trials are reviewed for their ability to minimize patient exposure to inferior experimental treatments. Examples of trials from the literature are discussed along with a computer simulation of a simple non-inferiority monitoring rule. Interim monitoring for non-inferiority trials is shown to substantially reduce the exposure of patients to inferior therapies when, in fact, the experimental treatment is inferior to the standard treatment. Interim monitoring rules typically used in superiority trials may be sub-optimal for non-inferiority trials, and may unnecessarily expose patients to inferior therapies. Examples of trials with inferior experimental arms and trials with sub-optimal monitoring rules are given. Appropriate interim monitoring of non-inferiority trials can reduce the exposure of patients to inferior therapies when the experimental treatment is inferior to the standard treatment.

Assessing treatment efficacy in the subset of responders in a randomized clinical trial.

BACKGROUND: Durability of response is a clinically relevant dimension of the treatment effect in randomized clinical trials; it is often measured by comparing among the responders the duration of response between the treatment arms. However, since the comparison groups are defined by response (a post-randomization event), it is subject to analysis-by-responder bias, especially if the proportion of responders differs between the arms. METHODS: Two simple methods are developed that use tumor shrinkage measurements in order to lessen analysis-by-responder bias by generating more comparable patient subsets in the control and experimental arms of the trial. These subsets are then used to estimate between-arm differences in response duration. In the subtraction method, responding patients with the least tumor shrinkage in the treatment arm with more responders are removed from the patient subset for that arm. In the addition method, non-responding patients with the most tumor shrinkage in the treatment arm with fewer responders are added to the patient subset for that arm. In both methods, the numbers of patients subtracted or added are such that the proportion of patients in the modified patient subset is the same as the proportion of responders in the other treatment arm. RESULTS: The methods are demonstrated on a hypothetical dataset where they are shown to eliminate analysis-by-responder bias, and on two published analyses of randomized trials that compared the duration of response between the treatment arms. CONCLUSIONS: The proposed methods can lessen the analysis-by-responder bias. These methods to compare duration of response between treatment arms may provide a useful exploratory analysis to measure treatment efficacy among responders.

Statistical controversies in clinical research: assessing pathologic complete response as a trial-level surrogate end point for early-stage breast cancer.

BACKGROUND: A trial-level surrogate end point for a randomized clinical trial may allow assessment of the relative benefits of the treatment to be performed at an earlier time point and potentially with a smaller sample size. However, determining whether an end point is a reliable trial-level surrogate based on results of previous trials is not straightforward. The question of trial-level surrogacy is easily confused with the question of individual-level surrogacy, and this confusion can lead to controversy. A recent example concerns the evaluation of pathologic complete response (pCR) as a surrogate for event-free survival (EFS) and overall survival (OS) in early-stage breast cancer. MATERIALS AND METHODS: The differences between individual-level surrogacy (i.e. for patients receiving a specific treatment, the surrogate end point predicts the definitive end point) and trial-level surrogacy (the results of the trial for the surrogate end point predict the results of the trial for the definitive end point) are discussed. Trial-level data used in two previous meta-analyses evaluating pCR as a trial-level surrogate for EFS and OS were re-analyzed using methods that appropriately account for the variability in pCR rates as well as the variability in the hazard ratios for EFS and OS. RESULTS: There is no evidence that pCR is a trial-level surrogate for EFS or OS, nor is there evidence that pCR could be used reliably to screen out nonpromising agents from further drug development. CONCLUSIONS: At present, neoadjuvant RCTs should continue to follow patients to observe EFS and OS to assess clinical benefit, and they should be designed with sufficient sample size to reliably assess EFS. However, one cannot rule out the possibility that future meta-analyses involving more trials and in which the patient population or class of treatments is restricted could suggest the validity of pCR as a trial-level surrogate for EFS or OS in some focused settings.

Evaluation of chemoresponse assays as predictive markers.

BACKGROUND: A chemoresponse assay that can be used to predict which patients will respond to which drugs would be useful in directing treatment. Two new analytic methods to assess the predictive ability of a chemoresponse assay have been proposed by Tian et al. METHODS: Three examples in which a hypothetical assay has no predictive ability are considered to evaluate the properties of the proposed analytic methods. RESULTS: For these specific examples, the proposed methods incorrectly suggest that the assay is predictive. CONCLUSIONS: The examples presented here demonstrate that it can be challenging to evaluate the predictive value of a chemoresponse assay.

Evaluation of the contribution of randomised cancer clinical trials evaluating agents without documented single-agent activity.

BACKGROUND: With the development of targeted agents, the approach to combination cancer therapy has evolved to focus on identifying ways in which pathway inhibition by one agent may enhance the activity of other agents. In theory, this implies that under this new paradigm, agents are no longer required to show single-agent activity, as the pathway inhibited by the targeted agent may only have a therapeutic effect when given with other agents. This raises the question of the extent to which anticancer agents without single-agent activity can contribute to effective combination regimens. PATIENTS AND METHODS: We reviewed outcomes of randomised phase 2 combination trials sponsored by the National Cancer Institute Cancer Therapy Evaluation Program that were activated in 2008 to 2017 and noted the single-agent activity of the experimental agents. RESULTS: Fifty-three trials were identified, and 50 had available results: 7 (14%), 15 (30%) and 28 (56%) had experimental agents with single-agent activity classified as active, inactive and indeterminate, respectively. Thirteen per cent (95% CI=1.7% to 40.5%) of trials evaluating inactive agents and 11.6% (95% CI=3.9% to 25.1%) of trials evaluating agents without known single-agent activity (pooled inactive and indeterminate) were positive, compared with 42.9% (95% CI=9.9% to 81.6%) for agents with single-agent activity. CONCLUSIONS: Incorporating agents without documented single-agent activity into treatment regimens is unlikely to produce meaningful improvements in activity unless there is compelling biological rationale. This finding has important implications for the prioritisation of anticancer agents for combination testing, and for the allocation of clinical trial resources.

Analyzing health surveys for cancer-related objectives.

Large-scale health surveys conducted by government agencies record information on a large number of health-related variables. We review the use of these data for performing analyses that address cancer-related objectives. After describing the conduct of a large-scale health survey (the third National Health and Nutrition Examination Survey [NHANES III]), we discuss some of the issues involved in analyzing data collected in such a survey. In particular, the use of sample weights in the analysis and the importance of accounting for the complex survey design when estimating standard errors are discussed. Six applications are then presented that involve the following: 1) estimating demographic factors associated with snuff use, 2) estimating the association of type of health insurance with the probability of receiving a digital rectal examination, 3) estimating the association of body iron stores with the probability of later developing cancer, 4) estimating the changing rates of mammography screening in the United States between 1987 and 1992, 5) evaluating smoking and alcohol consumption as risk factors for digestive cancer by use of a population-based, case-control study, and 6) evaluating a randomized community-intervention experiment to encourage smoking cessation. These applications use data from the National Health Interview Survey, the NHANES I Epidemiologic Followup Study, the 1986 National Mortality Followback Survey, and the Community Intervention Trial for Smoking Cessation. The availability of public-use data files is discussed for surveys sponsored by the U.S. government that collect health-related information. We demonstrate that statistical methods and computer software are available for analyzing public-use data files of surveys to address different types of cancer-related objectives.

Selecting drug combinations based on total equivalent dose (dose intensity)

We describe a mathematical model for selecting cytotoxic drugs and dosages for a combination regimen based on the antitumor activities of the drugs given as single agents and their organ-specific maximum tolerated doses. The regimen defined maximizes an approximate measure of antitumor effect subject to constraints on combined toxicity. This approach does not assume that the underlying dose-response curve is steep; nor does it assume that maximally dose-intense regimens are clinically appropriate in all situations. Whether the identified regimen is superior to standard treatments should be determined by prospective, randomized clinical trials. Determining which drugs to combine and in what proportions to combine them offers combinatorially huge numbers of possibilities. The method described here offers one approach to identifying combinations worthy of evaluation in prospective trials.

Quality-of-life assessment in cancer treatment protocols: research issues in protocol development.

BACKGROUND: Interest in incorporating quality of life as an end point in clinical studies of cancer treatment has intensified in recent years. PURPOSE: We provide practical suggestions that will assist investigators considering including quality-of-life assessment in phase III therapeutic trials. METHODS: We discuss issues important in study planning, including quality-of-life definition, priority studies for quality-of-life assessment, eligibility requirements, and design. CONCLUSIONS: Many of the problems that quality-of-life studies have encountered, from protocol approval to data analysis, could be addressed and alleviated during protocol development. This discussion is intended to assist and stimulate investigators conducting research in this area.

Zoned immune suppression of lymph nodes draining malignant melanoma: histologic and immunohistologic studies.

Node-to-node heterogeneity of reaction and recognizable patterns of reaction in node groups draining melanoma were sought. Nodes from 72 patients undergoing lymphadenectomy for high-risk, primary melanoma (57) or node-spread melanoma (15) were accurately oriented to the nearest melanoma. Reactivity of paracortex, follicular areas, and sinuses was assessed on a 0-3+ scale. Reactivity was prominent in paracortex and sinuses but varied from node to node within node groups. Nodes nearest to tumor showed least reaction; nodes at intermediate distances from tumor were most reactive, while those farthest away showed mostly little reaction. Variation of nodal reaction that correlated with the node position relative to the nearest melanoma (zoned reaction) was seen in 92% of patients with nodal metastases of melanoma and in 64% of patients with primary malignant melanoma. Follicular and sinusoidal reactions showed no significant zoning. S-100 protein-positive paracortical dendritic cells (PDCs) in tumor-oriented nodes were quantified. PDCs were infrequent in nodes partly replaced by melanoma or located near to melanoma but were numerous in nodes located farther from tumor. Changes of nodal activity (relative stimulation or suppression) correlate with the distance of the node from the nearest deposit of primary or metastatic melanoma.

Variations in functional immunocompetence of individual tumor-draining lymph nodes in humans.

In studying the histology and immunohistology of tumor-draining lymph nodes, we observed that nodes close to tumor showed reduced paracortical activity relative to those further away. To assess whether this reduction was paralleled by alterations in the functional activity of nodes, we examined the immunocompetence of individually oriented tumor-draining lymph nodes. We assessed the unstimulated activity of lymph node lymphocytes and their responses to phytohemagglutinin, interleukin 2, and one-way alloantigenic stimulation (mixed lymphocyte reaction) in vitro. Regional nodes from patients with malignant melanoma or breast cancer were classified as proximal, intermediate, or distal relative to tumor. Node groups with and without tumor metastases in them were studied. Significant variations in [3H]thymidine uptake were noted with the unstimulated and stimulated lymphocytes of different lymph nodes from the same individual. Nodes near to tumor were less responsive than those located further away; some of the latter showed relative immunostimulation. Tumor-draining node groups thus demonstrated a nonrandom variation in the strength of reaction of individual nodes. There are zones of low and high lymph node reactivity, related to the position of each node relative to tumor. Tumor-derived immunosuppressive products and/or immunoregulation down-regulates lymph node functional activity, creating conditions that may permit the survival of tumor cells and the establishment of metastases. It is suggested that immunosuppression of nodes nearest to tumor may facilitate the early establishment of metastases.

Clinical correlations with chemosensitivities measured in a rapid thymidine incorporation assay.

A rapid assay for in vitro chemosensitivity testing measuring [3H]thymidine incorporation has been developed. Results of this assay correlate highly with chemosensitivities determined by the soft-agar clonogenic assay. A correlative study was carried out on 219 solid tumor specimens to assess the ability of the rapid assay to predict clinical response to antineoplastic therapy. One hundred forty-two of 219 tumors (65%) yielded chemosensitivity data. Of these, 33 were evaluable for in vitro-in vivo correlations. In vitro sensitivity (greater than or equal to 80% inhibition of thymidine uptake) was associated with clinical response in 6 of 13 patients. In vitro resistance was associated with progressive disease in 20 of 20 patients. The rapid assay offers several advantages over the soft-agar clonogenic assay, including higher success rate, avoidance of clumping artifact, shorter time course (5 days), and very low false-negative rate. Further refinement may be necessary, but the rapid assay appears to have potential for individualizing solid tumor chemotherapy.

Using the tolerable-dose diagram in the design of phase I combination chemotherapy trials.

PURPOSE: Choosing maximum-tolerated doses (MTDs) of agents to be administered in combination is more complicated than choosing the MTD for a single agent. This is because there are many combinations of doses that will be tolerated. We offer guidance in targeting specific MTD combinations, and suggest trial designs to achieve these targets. METHODS: A graphical method based on a simple, previously described mathematical model is used to guide the phase I trial design. The method involves constructing a tolerable-dose diagram, which displays the toxicities that are expected to occur at various dose combinations. The data required for the method are the single-agent toxicity profiles of the agents. Designs for combinations of taxol with fluorouracil, carboplatin, doxorubicin, cyclophosphamide, and cisplatin are used to demonstrate the methods. RESULTS: For all of the drugs considered here, leukopenia is dose-limiting or nearly dose-limiting. Consequently, no major improvement in total dose-intensity is achievable by combining these drugs. If leukopenia can be eliminated or substantially reduced by use of chemoprotective agents, then a cyclophosphamide/taxol combination appears promising from a dose-intensity point of view. For other combinations, the doses must be reduced from single-agent MTD levels. In the absence of biologic information such as concentrations needed for optimal modulation, it is suggested that single-agent MTDs be reduced approximately proportionately for each drug in the combination. CONCLUSION: Tolerable-dose diagrams are useful for planning phase I trials of combinations of agents. They can suggest which combinations are promising from a dose-intensity perspective, as well as dose-escalation schemes for combinations to be pursued for dose-intensity or other considerations.

Role of independent data-monitoring committees in randomized clinical trials sponsored by the National Cancer Institute.

PURPOSE: To describe the rationale for independent data monitoring committees (DMCs) for National Cancer Institute (NCI)-sponsored phase III cooperative group clinical trials. DESIGN: We review the necessity for interim monitoring of outcome data during the course of randomized clinical trials and summarize the reasons for establishing DMCs with requisite expertise and with appropriate independence from study investigators. RESULTS: The important components of the policy for cooperative group DMCs are described with a focus on the makeup of these bodies and on the complementary roles of study committee leadership and DMCs in protecting patient safety during the conduct of randomized clinical trials. CONCLUSION: The cooperative group DMCs that are independent of the study committees and that have the requisite expertise to examine accumulating data and to base decisions on monitoring guidelines that are specified in advance by the study committee provide a body able to protect patient safety, to protect the integrity of the clinical experiments on which patients have consented to participate, and to assure the public that conflicts of interest do not compromise either patient safety or trial integrity.

Clinical trial designs for cytostatic agents: are new approaches needed?

Preclinical data suggest that some new anticancer agents directed at novel targets demonstrate tumor growth inhibition but not tumor shrinkage. Such cytostatic agents may offer clinical benefits for patients in the absence of tumor shrinkage. In addition, lower doses of some of these agents may be just as effective as higher doses, implying that toxicity may not be an ideal end point for dose finding. Because of these factors, the sequence and design of traditional phase I, II, and III trials used for cytotoxic agents (which typically shrink tumors and in a dose-dependent manner) may not be appropriate for cytostatic agents. This article discusses options for modifying trial designs to accommodate cytostatic agents. Examples are given where these options have been tried or are currently being tried. Recommendations given for choosing among the trial designs depend on what is known preclinically about the agents (eg, does one have a validated and reproducible biologic end point that can be used to guide a dose escalation?), what is known about the patient population being studied (eg, does one have a well-documented historical progression-free survival rate at 1 year for comparison with the experience of the new agent?), and the numbers of agents and patients available for participation in trials. Planned and ongoing trials will test the utility of some of these new approaches.

Uniform approach to risk classification and treatment assignment for children with acute lymphoblastic leukemia.

PURPOSE: To define more uniform criteria for risk-based treatment assignment for children with acute lymphoblastic leukemia (ALL), the Cancer Therapy Evaluation Program (CTEP) of the National Cancer Institute (NCI) sponsored a workshop in September 1993. Participants included representatives from the Childrens Cancer Group (CCG), Pediatric Oncology Group (POG), Dana-Farber Cancer Institute (DFCI), St Jude Children's Research Hospital (SJCRH), and the CTEP. METHODS: Workshop participants presented and reviewed data from ALL clinical trials, using weighted averages to combine outcome data from different groups. RESULTS: For patients with B-precursor (ie, non-T, non-B) ALL, the standard-risk category (4-year event-free survival [EFS] rate, approximately 80%) will include patients 1 to 9 years of age with a WBC count at diagnosis less than 50,000/microL. The remaining patients will be classified as having high-risk ALL (4-year EFS rate, approximately 65%). For patients with T-cell ALL, different treatment strategies have yielded different conclusions concerning the prognostic significance of T-cell immunophenotype. Therefore, some groups/institutions will classify patients with T-cell ALL as high risk, while others will assign risk for patients with T-cell ALL based on the uniform age/WBC count criteria. Workshop participants agreed that the risk category of a patient may be modified by prognostic factors in addition to age and WBC count criteria, and that a common set of prognostic factors should be uniformly obtained, including DNA index (DI), cytogenetics, early response to treatment (eg, day-14 bone marrow), immunophenotype, and CNS status. CONCLUSIONS: The more uniform approach to risk-based treatment assignment and to collection of specific prognostic factors should increase the efficiency of future ALL clinical research.

Quantification of natural cytotoxicity by human lymphocyte subpopulations isolated by density: heterogeneity of the effector cells.

Natural cell mediated cytotoxicity has been expressed as percent cytotoxicity, as the slope of the titration curve obtained by testing different effector: target cell ratios, and as lytic units. Objections can be raised to each method as used. The present report involves the study of cytotoxicity by subpopulations of lymphocytes obtained by Percoll density gradient centrifugation. The subpopulations vary greatly in cytotoxic activity, making accurate comparisons by traditional means difficult. A method was therefore developed for making objective comparisons between activities of subpopulations of lymphocytes. Cytotoxicity titration curves, which were sigmoidal on linear-linear plots, were found to best fit the sigmoidal curves described by the Von Krogh equation. A best fitting scale family of curves having the identical maximum cytotoxicity and shape parameter was fitted simultaneously to cytotoxicity measurements obtained by titrating all subpopulations of effector cells obtained from each patient. Values expressing relative cytotoxicities were obtained from the ratios of the scale parameters of the different curves or by obtaining lytic units from the fitted curves. In addition to the enriched natural cytotoxic activity found among the lighter cells obtained by Percoll density gradient centrifugation, activity was also increased among cells sedimenting at the very bottom of the gradient. This was found in subpopulations of cells obtained from 16/21 gradients in tests against K562 and in subpopulations obtained from both of 2 gradients in tests against Daudi cells. These findings are consistent with the existence of at least 2 subpopulations of lymphocytes which can mediate natural cytotoxicity, and which are separable by density.

The use of three baseline values in intervention studies: application to evaluation of immune modulation therapies.

In studying the effects of a treatment intervention on immunological parameters, we have found that three baseline values are a practical number to obtain on each patient. Three baseline values 1) increase the chances of detecting a statistically significant effect of the intervention, 2) provide an assessment of the daily variability of the assay and patients, and 3) enable the identification of individual patients who demonstrate significant changes associated with the intervention.

Morphometric analysis of the homogeneity of lymphoid cell populations in fine-needle aspiration cytology smears.

Quantitative morphometry was used to evaluate the homogeneity of lymphoid cell populations in fine-needle aspiration cytology smears of lymph nodes from eight non-Hodgkin's lymphomas and four benign lymphoid proliferations. Using the standard deviation of log-transformed nuclear areas as the measure of lymphoid cell homogeneity, diffuse small cell lymphomas were found to be significantly more homogeneous than diffuse large cell lymphomas or benign lymphoid proliferations, but there was no significant difference in the homogeneity of the latter two groups. Two of four large cell lymphomas showed a bimodal distribution of nuclear areas. Methodologic issues--including the random selection of cells, the number of cells that should be counted in each case, the significance of microscopic field-to-field variation in this setting, and the significance of drift in observations during the examination of individual cases--are discussed.

Immunological responses of aging Japanese A-bomb survivors.

Immune response parameters were studied on 1341 A-bomb survivors residing in Hiroshima, Japan. Mononuclear cells were isolated from venous blood and tested for interleukin-2 production; lymphocytes were purified and tested for natural killer (NK) cell activity and interferon (IFN) production; and serum was tested for IFN and circulating immune complex (CIC) levels. Statistical analyses were performed for each type of assay using a linear models procedure including sex, age at the time of the bomb, radiation exposure, all the interaction variables, and the categorical variable day-of-assay in the model. The findings showed that (1) none of the immunologic variables were significantly affected by radiation exposure; (2) NK activity and CIC levels were positively associated with age; and (3) NK activity was on average higher for males than females. The data exemplify the difficulty in reaching firm conclusions concerning associations with radiation exposure when the dependent variable exhibits a large degree of interindividual and day-of-assay variability.