Current drug development and trial designs in neuro-oncology: report from the first American Society of Clinical Oncology and Society for Neuro-Oncology Clinical Trials Conference.

Successful drug development for people with cancers of the CNS has been challenging. There are multiple barriers to successful drug development including biological factors, rarity of the disease, and ineffective use of clinical trials. Based upon a series of presentations at the First Central Nervous System Clinical Trials Conference hosted by the American Society of Clinical Oncology and the Society for Neuro-Oncology, we provide an overview on drug development and novel trial designs in neuro-oncology. This Review discusses the challenges of therapeutic development in neuro-oncology and proposes strategies to improve the drug discovery process by enriching the pipeline of promising therapies, optimising trial design, incorporating biomarkers, using external data, and maximising efficacy and reproducibility of clinical trials.

Prognostic value of interim FDG-PET in diffuse large cell lymphoma: results from the CALGB 50303 Clinical Trial.

As part of a randomized, prospective clinical trial in large cell lymphoma, we conducted serial fluorodeoxyglucose positron emission tomography (FDG-PET) at baseline, after 2 cycles of chemotherapy (interim PET [i-PET]), and at end of treatment (EoT) to identify biomarkers of response that are predictive of remission and survival. Scans were interpreted in a core laboratory by 2 imaging experts, using the visual Deauville 5-point scale (5-PS), and by calculating percent change in FDG uptake (change in standardized uptake value [ΔSUV]). Visual scores of 1 through 3 and ΔSUV ≥66% were prospectively defined as negative. Of 524 patients enrolled in the parent trial, 169 agreed to enroll in the PET substudy and 158 were eligible for final analysis. In this selected population, all had FDG-avid disease at baseline; by 5-PS, 55 (35%) remained positive on i-PET and 28 (18%) on EoT PET. Median ΔSUV on i-PET was 86.2%. With a median follow-up of 5 years, ΔSUV, as continuous variable, was associated with progression-free survival (PFS) (hazard ratio [HR] = 0.99; 95% confidence interval [CI], 0.97-1.00; P = .02) and overall survival (OS) (HR, 0.98; 95% CI, 0.97-0.99; P = .03). ΔSUV ≥66% was predictive of OS (HR, 0.31; 95% CI, 0.11-0.85; P = .02) but not PFS (HR, 0.47; 95% CI, 0.19-1.13; P = .09). Visual 5-PS on i-PET did not predict outcome. ΔSUV, but not visual analysis, on i-PET predicted OS in DLBCL, although the low number of events limited the statistical analysis. These data may help guide future clinical trials using PET response-adapted therapy. This trial was registered at www.clinicaltrials.gov as #NCT00118209.

Leveraging external data in the design and analysis of clinical trials in neuro-oncology.

Integration of external control data, with patient-level information, in clinical trials has the potential to accelerate the development of new treatments in neuro-oncology by contextualising single-arm studies and improving decision making (eg, early stopping decisions). Based on a series of presentations at the 2020 Clinical Trials Think Tank hosted by the Society of Neuro-Oncology, we provide an overview on the use of external control data representative of the standard of care in the design and analysis of clinical trials. High-quality patient-level records, rigorous methods, and validation analyses are necessary to effectively leverage external data. We review study designs, statistical methods, risks, and potential distortions in using external data from completed trials and real-world data, as well as data sources, data sharing models, ongoing work, and applications in glioblastoma.

A clinical calculator to predict disease outcomes in women with hormone receptor-positive advanced breast cancer treated with first-line endocrine therapy.

PURPOSE: Endocrine therapy (ET) is an effective strategy to treat hormone receptor-positive, human epidermal growth factor receptor 2-negative (HR+/HER2-) advanced breast cancer (ABC) but nearly all patients eventually progress. Our goal was to develop and validate a web-based clinical calculator for predicting disease outcomes in women with HR+ABC who are candidates for receiving first-line single-agent ET. METHODS: The meta-database comprises 891 patient-level data from the control arms of five contemporary clinical trials where patients received first-line single-agent ET (either aromatase inhibitor or fulvestrant) for ABC. Risk models were constructed for predicting 24-months progression-free survival (PFS-24) and 24-months overall survival (OS-24). Final models were internally validated for calibration and discrimination using ten-fold cross-validation. RESULTS: Higher number of sites of metastases, measurable disease, younger age, lower body mass index, negative PR status, and prior endocrine therapy were associated with worse PFS. Final PFS and OS models were well-calibrated and associated with cross-validated time-dependent area under the curve (AUC) of 0.61 and 0.62, respectively. CONCLUSIONS: The proposed ABC calculator is internally valid and can accurately predict disease outcomes. It may be used to predict patient prognosis, aid planning of first-line treatment strategies, and facilitate risk stratification for future clinical trials in patients with HR+ABC. Future validation of the proposed models in independent patient cohorts is warranted.

A clinical calculator to predict disease outcomes in women with triple-negative breast cancer.

PURPOSE: Triple-negative breast cancer (TNBC) is the most aggressive subtype of breast cancer, characterized by substantial risks of early disease recurrence and mortality. We constructed and validated clinical calculators for predicting recurrence-free survival (RFS) and overall survival (OS) for TNBC. METHODS: Data from 605 women with centrally confirmed TNBC who underwent primary breast cancer surgery at Mayo Clinic during 1985-2012 were used to train risk models. Variables included age, menopausal status, tumor size, nodal status, Nottingham grade, surgery type, adjuvant radiation therapy, adjuvant chemotherapy, Ki67, stromal tumor-infiltrating lymphocytes (sTIL) score, and neutrophil-to-lymphocyte ratio (NLR). Final models were internally validated for calibration and discrimination using ten-fold cross-validation and compared with their base-model counterparts which include only tumor size and nodal status. Independent external validation was performed using data from 478 patients diagnosed with stage II/III invasive TNBC during 1986-1992 in the British Columbia Breast Cancer Outcomes Unit database. RESULTS: Final RFS and OS models were well calibrated and associated with C-indices of 0.72 and 0.73, as compared with 0.64 and 0.62 of the base models (p < 0.001). In external validation, the discriminant ability of the final models was comparable to the base models (C-index: 0.59-0.61). The RFS model demonstrated greater accuracy than the base model both overall and within patient subgroups, but the advantages of the OS model were less profound. CONCLUSIONS: This TNBC clinical calculator can be used to predict patient outcomes and may aid physician's communication with TNBC patients regarding their long-term disease outlook and planning treatment strategies.

Criteria for the use of omics-based predictors in clinical trials.

The US National Cancer Institute (NCI), in collaboration with scientists representing multiple areas of expertise relevant to 'omics'-based test development, has developed a checklist of criteria that can be used to determine the readiness of omics-based tests for guiding patient care in clinical trials. The checklist criteria cover issues relating to specimens, assays, mathematical modelling, clinical trial design, and ethical, legal and regulatory aspects. Funding bodies and journals are encouraged to consider the checklist, which they may find useful for assessing study quality and evidence strength. The checklist will be used to evaluate proposals for NCI-sponsored clinical trials in which omics tests will be used to guide therapy.

Power estimation in biomarker studies where events are already observed.

Background The clinical utility of a new biomarker should ideally be established in a prospective randomized clinical trial. However, such trials are not always practical. As such, it is common for investigators to identify promising biomarkers using archived specimens and clinical data collected from previously completed therapeutic trials. Simon et al. defined such biomarker studies as prospective-retrospective studies and proposed specific conditions to satisfy for such evaluations to be more than hypothesis generating. One condition they proposed is that archived tissues must be available on a sufficiently large number of patients from the pivotal trials to ensure adequately powered analyses. Purpose The purpose of this article is to provide a new perspective on how to estimate power for assessing the prognostic and predictive values of a single binary biomarker in prospective-retrospective biomarker studies. Computer programs are provided to facilitate the use of these methods in practice. Methods The proposed methods utilize additional information that becomes available during the course of the treatment trial including sample size, accrual time, additional follow-up time, and the observed number of events at time of biomarker analysis. These methods involve solving for the exponential hazard rates that give rise to the event numbers that are consistent with those observed while satisfying other design parameter constraints. Conclusion Simon et al. proposed a new paradigm for biomarker design, conduct, analysis, and evaluation in prospective-retrospective studies that offers a more efficient alternative than fully prospective biomarker studies. As a general rule, they suggest that samples from at least two-thirds of the patients be available for analysis. In this article, I expand on this issue and provide a methodological tool useful for estimating study power in prospective-retrospective biomarker studies. It is my hope that these incremental efforts to improve the quality and statistical rigor in biomarker studies will hasten the introduction of useful tumor biomarkers into clinical practice.

An international study to increase concordance in Ki67 scoring.

Although an important biomarker in breast cancer, Ki67 lacks scoring standardization, which has limited its clinical use. Our previous study found variability when laboratories used their own scoring methods on centrally stained tissue microarray slides. In this current study, 16 laboratories from eight countries calibrated to a specific Ki67 scoring method and then scored 50 centrally MIB-1 stained tissue microarray cases. Simple instructions prescribed scoring pattern and staining thresholds for determination of the percentage of stained tumor cells. To calibrate, laboratories scored 18 'training' and 'test' web-based images. Software tracked object selection and scoring. Success for the calibration was prespecified as Root Mean Square Error of scores compared with reference <0.6 and Maximum Absolute Deviation from reference <1.0 (log2-transformed data). Prespecified success criteria for tissue microarray scoring required intraclass correlation significantly >0.70 but aiming for observed intraclass correlation ≥0.90. Laboratory performance showed non-significant but promising trends of improvement through the calibration exercise (mean Root Mean Square Error decreased from 0.6 to 0.4, Maximum Absolute Deviation from 1.6 to 0.9; paired t-test: P=0.07 for Root Mean Square Error, 0.06 for Maximum Absolute Deviation). For tissue microarray scoring, the intraclass correlation estimate was 0.94 (95% credible interval: 0.90-0.97), markedly and significantly >0.70, the prespecified minimum target for success. Some discrepancies persisted, including around clinically relevant cutoffs. After calibrating to a common scoring method via a web-based tool, laboratories can achieve high inter-laboratory reproducibility in Ki67 scoring on centrally stained tissue microarray slides. Although these data are potentially encouraging, suggesting that it may be possible to standardize scoring of Ki67 among pathology laboratories, clinically important discrepancies persist. Before this biomarker could be recommended for clinical use, future research will need to extend this approach to biopsies and whole sections, account for staining variability, and link to outcomes.

Two-stage adaptive cutoff design for building and validating a prognostic biomarker signature.

Cancer biomarkers are frequently evaluated using archived specimens collected from previously conducted therapeutic trials. Routine collection and banking of high quality specimens is an expensive and time-consuming process. Therefore, care should be taken to preserve these precious resources. Here, we propose a novel two-stage adaptive cutoff design that affords the possibility to stop the biomarker study early if an evaluation of the model performance is unsatisfactory at an early stage, thereby allowing one to preserve the remaining specimens for future research. In addition, our design integrates important elements necessary to meet statistical rigor and practical demands for developing and validating a prognostic biomarker signature, including maintaining strict separation between the datasets used to build and evaluate the model and producing a locked-down signature to facilitate future validation. We conduct simulation studies to evaluate the operating characteristics of the proposed design. We show that under the null hypothesis when the model performance is deemed undesirable, the proposed design maintains type I error at the nominal level, has high probabilities of terminating the study early, and results in substantial savings in specimens. Under the alternative hypothesis, power is generally high when the total sample size and the targeted degree of improvement in prediction accuracy are reasonably large. We illustrate the use of the procedure with a dataset in patients with diffuse large-B-cell lymphoma. The practical aspects of the proposed designs are discussed. Published 2014. This article is a U.S. Government work and is in the public domain in the USA.

Comparison of confidence interval methods for an intra-class correlation coefficient (ICC).

BACKGROUND: The intraclass correlation coefficient (ICC) is widely used in biomedical research to assess the reproducibility of measurements between raters, labs, technicians, or devices. For example, in an inter-rater reliability study, a high ICC value means that noise variability (between-raters and within-raters) is small relative to variability from patient to patient. A confidence interval or Bayesian credible interval for the ICC is a commonly reported summary. Such intervals can be constructed employing either frequentist or Bayesian methodologies. METHODS: This study examines the performance of three different methods for constructing an interval in a two-way, crossed, random effects model without interaction: the Generalized Confidence Interval method (GCI), the Modified Large Sample method (MLS), and a Bayesian method based on a noninformative prior distribution (NIB). Guidance is provided on interval construction method selection based on study design, sample size, and normality of the data. We compare the coverage probabilities and widths of the different interval methods. RESULTS: We show that, for the two-way, crossed, random effects model without interaction, care is needed in interval method selection because the interval estimates do not always have properties that the user expects. While different methods generally perform well when there are a large number of levels of each factor, large differences between the methods emerge when the number of one or more factors is limited. In addition, all methods are shown to lack robustness to certain hard-to-detect violations of normality when the sample size is limited. CONCLUSIONS: Decision rules and software programs for interval construction are provided for practical implementation in the two-way, crossed, random effects model without interaction. All interval methods perform similarly when the data are normal and there are sufficient numbers of levels of each factor. The MLS and GCI methods outperform the NIB when one of the factors has a limited number of levels and the data are normally distributed or nearly normally distributed. None of the methods work well if the number of levels of a factor are limited and data are markedly non-normal. The software programs are implemented in the popular R language.

Leveraging external control data in the design and analysis of neuro-oncology trials: Pearls and perils.

BACKGROUND: Randomized controlled trials have been the gold standard for evaluating medical treatments for many decades but they are often criticized for requiring large sample sizes. Given the urgent need for better therapies for glioblastoma, it has been argued that data collected from patients treated with the standard regimen can provide high-quality external control data to supplement or replace concurrent control arm in future glioblastoma trials. METHODS: In this article, we provide an in-depth appraisal of the use of external control data in the context of neuro-oncology trials. We describe several clinical trial designs with particular attention to how external information is utilized and address common fallacies that may lead to inappropriate adoptions of external control data. RESULTS: Using 2 completed glioblastoma trials, we illustrate the use of an assessment tool that lays out a blueprint for assembling a high-quality external control data set. Using statistical simulations, we draw caution from scenarios where these approaches can fall short on controlling the type I error rate. CONCLUSIONS: While this approach may hold promise in generating informative data in certain settings, this sense of optimism should be tampered with a healthy dose of skepticism due to a myriad of design and analysis challenges articulated in this review. Importantly, careful planning is key to its successful implementation.

Criteria for the use of omics-based predictors in clinical trials: explanation and elaboration.

High-throughput 'omics' technologies that generate molecular profiles for biospecimens have been extensively used in preclinical studies to reveal molecular subtypes and elucidate the biological mechanisms of disease, and in retrospective studies on clinical specimens to develop mathematical models to predict clinical endpoints. Nevertheless, the translation of these technologies into clinical tests that are useful for guiding management decisions for patients has been relatively slow. It can be difficult to determine when the body of evidence for an omics-based test is sufficiently comprehensive and reliable to support claims that it is ready for clinical use, or even that it is ready for definitive evaluation in a clinical trial in which it may be used to direct patient therapy. Reasons for this difficulty include the exploratory and retrospective nature of many of these studies, the complexity of these assays and their application to clinical specimens, and the many potential pitfalls inherent in the development of mathematical predictor models from the very high-dimensional data generated by these omics technologies. Here we present a checklist of criteria to consider when evaluating the body of evidence supporting the clinical use of a predictor to guide patient therapy. Included are issues pertaining to specimen and assay requirements, the soundness of the process for developing predictor models, expectations regarding clinical study design and conduct, and attention to regulatory, ethical, and legal issues. The proposed checklist should serve as a useful guide to investigators preparing proposals for studies involving the use of omics-based tests. The US National Cancer Institute plans to refer to these guidelines for review of proposals for studies involving omics tests, and it is hoped that other sponsors will adopt the checklist as well.

Development of omics-based clinical tests for prognosis and therapy selection: the challenge of achieving statistical robustness and clinical utility.

BACKGROUND: Many articles have been published in biomedical journals reporting on the development of prognostic and therapy-guiding biomarkers or predictors developed from high-dimensional data generated by omics technologies. Few of these tests have advanced to routine clinical use. PURPOSE: We discuss statistical issues in the development and evaluation of prognostic and therapy-guiding biomarkers and omics-based tests. METHODS: Concepts relevant to the development and evaluation of prognostic and therapy-guiding clinical tests are illustrated through discussion and examples. Some differences between statistical approaches for test evaluation and therapy evaluation are explained. The additional complexities introduced in the evaluation of omics-based tests are highlighted. RESULTS: Distinctions are made between clinical validity of a test and clinical utility. To establish clinical utility for prognostic tests, it is explained why absolute risk should be evaluated in addition to relative risk measures. The critical role of an appropriate control group is emphasized for evaluation of therapy-guiding tests. Common pitfalls in the development and evaluation of tests generated from high-dimensional omics data such as model overfitting and inappropriate methods for test performance evaluation are explained, and proper approaches are suggested. LIMITATIONS: The cited references do not comprise an exhaustive list of useful references on this topic, and a systematic review of the literature was not performed. Instead, a few key points were highlighted and illustrated with examples drawn from the oncology literature. CONCLUSIONS: Approaches for the development and statistical evaluation of clinical tests useful for predicting prognosis and selecting therapy differ from standard approaches for therapy evaluation. Proper evaluation requires an understanding of the clinical setting and what information is likely to influence clinical decisions. Specialized expertise relevant to building mathematical predictor models from high-dimensional data is helpful to avoid common pitfalls in the development and evaluation of omics-based tests.

Molecular Profiling in Neuro-Oncology: Where We Are, Where We're Heading, and How We Ensure Everyone Can Come Along.

Advances in molecular profiling have led to improved understanding of glioma heterogeneity. Results have been used to inform diagnostic classification and targeted treatment strategies. Validation of these tests is necessary in the development of biomarkers that can aid in treatment decision, allowing for personalized medicine in neuro-oncologic diseases. Although not all populations have benefitted equally from awareness of and access to testing, opportunities arise regarding incorporating this testing into the standard of care for patients with glioma.

Targeted gene expression profiling predicts meningioma outcomes and radiotherapy responses.

Surgery is the mainstay of treatment for meningioma, the most common primary intracranial tumor, but improvements in meningioma risk stratification are needed and indications for postoperative radiotherapy are controversial. Here we develop a targeted gene expression biomarker that predicts meningioma outcomes and radiotherapy responses. Using a discovery cohort of 173 meningiomas, we developed a 34-gene expression risk score and performed clinical and analytical validation of this biomarker on independent meningiomas from 12 institutions across 3 continents (N = 1,856), including 103 meningiomas from a prospective clinical trial. The gene expression biomarker improved discrimination of outcomes compared with all other systems tested (N = 9) in the clinical validation cohort for local recurrence (5-year area under the curve (AUC) 0.81) and overall survival (5-year AUC 0.80). The increase in AUC compared with the standard of care, World Health Organization 2021 grade, was 0.11 for local recurrence (95% confidence interval 0.07 to 0.17, P < 0.001). The gene expression biomarker identified meningiomas benefiting from postoperative radiotherapy (hazard ratio 0.54, 95% confidence interval 0.37 to 0.78, P = 0.0001) and suggested postoperative management could be refined for 29.8% of patients. In sum, our results identify a targeted gene expression biomarker that improves discrimination of meningioma outcomes, including prediction of postoperative radiotherapy responses.

Two-Stage Adaptive Design for Prognostic Biomarker Signatures with a Survival Endpoint.

Cancer biomarker discoveries typically involve utilizing patient specimens. In practice, there is often strong desire to preserve high quality biospecimens for studies that are most likely to yield useful information. Previously, we proposed a two-stage adaptive design for binary endpoints which terminates the biomarker study in a futility interim if the model performance is unsatisfactory. In this work, we extend the two-stage design framework to accommodate time-to-event endpoints. The first stage of the procedure involves testing whether the measure of discrimination for survival models (C-index) exceeds a pre-specified threshold. We describe the computation of cross-validated C-index and evaluation of the statistical significance using re-sampling techniques. The second stage involves an independent model validation. Our simulation studies show that under the null hypothesis, the proposed design maintains type I error at the nominal level and has high probabilities of terminating the study early. Under the alternative hypothesis, power of the design is a function of the true event proportion, the sample size, and the targeted improvement in the discriminant measure. We apply the method to design of a prognostic biomarker study in patients with triple-negative breast cancer. Some practical aspects of the proposed method are discussed.

Joint Final Report of EORTC 26951 and RTOG 9402: Phase III Trials With Procarbazine, Lomustine, and Vincristine Chemotherapy for Anaplastic Oligodendroglial Tumors.

Clinical trials frequently include multiple end points that mature at different times. The initial report, typically based on the basis of the primary end point, may be published when key planned co-primary or secondary analyses are not yet available. Clinical Trial Updates provide an opportunity to disseminate additional results from studies, published in JCO or elsewhere, for which the primary end point has already been reported.Anaplastic oligodendroglial tumors (AOTs) are chemotherapy-sensitive brain tumors. We report the final very long-term survival results from European Organization for the Research and Treatment of Cancer 26951 and Radiation Therapy Oncology Group 9402 phase III trials initiated in 1990s, which both studied radiotherapy with/without neo/adjuvant procarbazine, lomustine, and vincristine (PCV) for newly diagnosed anaplastic oligodendroglial tumors. The median follow-up duration in both was 18-19 years. For European Organization for the Research and Treatment of Cancer 26951, median, 14-year, and probable 20-year overall survival rates without versus with PCV were 2.6 years, 13.4%, and 10.1% versus 3.5 years, 25.1%, and 16.8% (N = 368 overall; hazard ratio [HR] 0.78; 95% CI, 0.63 to 0.98; P = .033), with 1p19q codeletion 9.3 years, 26.2%, and 13.6% versus 14.2 years, 51.0%, and 37.1% (n = 80; HR 0.60; 95% CI, 0.35 to 1.03; P = .063), respectively. For Radiation Therapy Oncology Group 9402, analogous results were 4.8 years, 16.5%, and 11.2% versus 4.8 years, 29.1%, and 24.6% (N = 289 overall; HR 0.79; 95% CI, 0.61 to 1.03; P = .08), with codeletion 7.3 years, 25.0%, and 14.9% versus 13.2 years, 46.1%, and 37% (n = 125; HR 0.61; 95% CI, 0.40 to 0.94; P = .02), respectively. With that, the studies show similar long-term survival even without tumor recurrence in a significant proportion of patients after first-line treatment with radiotherapy/PCV.

CALGB 40603 (Alliance): Long-Term Outcomes and Genomic Correlates of Response and Survival After Neoadjuvant Chemotherapy With or Without Carboplatin and Bevacizumab in Triple-Negative Breast Cancer.

PURPOSE: CALGB 40603 (NCT00861705), a 2 × 2 randomized phase II trial, demonstrated that adding carboplatin or bevacizumab to weekly paclitaxel (wP) followed by doxorubicin and cyclophosphamide significantly increased the pathologic complete response (pCR) rate in stage II-III triple-negative breast cancer. We now report long-term outcomes (LTOs) and correlative science end points. PATIENTS AND METHODS: The Kaplan-Meier method was used to estimate LTOs in 443 patients who initiated study treatment. Log-rank tests and Cox proportional hazards models evaluated the impact of clinical characteristics, pathologic response, calculated residual cancer burden (RCB) in patients with residual disease (RD), treatment assignment, and dose delivery during wP on LTOs, including event-free survival (EFS). Genomic predictors of treatment response and outcomes were assessed on pretreatment tumor samples by mRNA sequencing. RESULTS: Among baseline characteristics, only the clinical stage was associated with LTOs. At a median follow-up of 7.9 years, LTOs were not significantly improved with either carboplatin or bevacizumab, overall or in patients with basal-like subtype cancers by genomic analysis. Patients with pCR (n = 205, 46.3%) had significantly higher 5-year EFS (85.5% v 56.6%, log-rank P < .0001) and overall survival (87.9% v 63.4%, P < .0001) rates compared with patients with RD, even those with RCB class I. Among clinical and genomic features, evidence of immune activation, including tumor-infiltrating lymphocytes and low B-cell receptor evenness, was associated with pCR and improved EFS. CONCLUSION: Despite higher pCR rates, neither carboplatin nor bevacizumab appeared to improve LTOs although the study was not powered to assess these secondary end points. pCR was associated with superior LTOs even when compared with minimal RD. Markers of immune activation in pretreatment tumor biopsies were independently associated with higher pCR rates and improved survival.

Practical modifications to the time-to-event continual reassessment method for phase I cancer trials with fast patient accrual and late-onset toxicities.

The goal of phase I cancer trials is to determine the highest dose of a treatment regimen with an acceptable toxicity rate. Traditional designs for phase I trials, such as the Continual Reassessment Method (CRM) and the 3 + 3 design, require each patient or a cohort of patients to be fully evaluated for the dose-limiting toxicity (DLT) before new patients can be enrolled. As such, the trial duration may be prohibitively long. The Time-to-Event Continual Reassessment Method (TITE-CRM, Cheung and Chappell, 2000) circumvents this limitation by allowing staggered patient accrual without the need for complete DLT follow-up of previously treated patients. However, in the setting of fast patient accrual and late-onset toxicities, the TITE-CRM results in overly aggressive dose escalation and exposes a considerable number of patients to toxic doses. We examine a modification to the TITE-CRM proposed by the original TITE-CRM creator and propose an alternative approach useful in this setting by incorporating an accrual suspension rule. A simulation study designed based on a neuro-oncology trial indicates that the modified methods provide a much improved degree of safety than the TITE-CRM while maintaining desirable design accuracy. The practical aspects of the proposed designs are discussed. The modifications presented are useful when planning phase I trials involving chemoradiation therapy.

Statistical Considerations in the Evaluation of Continuous Biomarkers.

Discovery of biomarkers has been steadily increasing over the past decade. Although a plethora of biomarkers has been reported in the biomedical literature, few have been sufficiently validated for broader clinical applications. One particular challenge that may have hindered the adoption of biomarkers into practice is the lack of reproducible biomarker cut points. In this article, we attempt to identify some common statistical issues related to biomarker cut point identification and provide guidance on proper evaluation, interpretation, and validation of such cut points. First, we illustrate how discretization of a continuous biomarker using sample percentiles results in significant information loss and should be avoided. Second, we review the popular "minimal-P-value" approach for cut point identification and show that this method results in highly unstable P values and unduly increases the chance of significant findings when the biomarker is not associated with outcome. Third, we critically review a common analysis strategy by which the selected biomarker cut point is used to categorize patients into different risk categories and then the difference in survival curves among these risk groups in the same dataset is claimed as the evidence supporting the biomarker's prognostic strength. We show that this method yields an exaggerated P value and overestimates the prognostic impact of the biomarker. We illustrate that the degree of the optimistic bias increases with the number of variables being considered in a risk model. Finally, we discuss methods to appropriately ascertain the additional prognostic contribution of the new biomarker in disease settings where standard prognostic factors already exist. Throughout the article, we use real examples in oncology to highlight relevant methodologic issues, and when appropriate, we use simulations to illustrate more abstract statistical concepts.