Number Of Dose Levels Research Articles

A review of dose escalation for FDA-approved products treating solid tumors and hematological malignancies in first-in-human trials.

First-in-human (FIH) dose-escalation trials on oncology should prioritize safety and emphasize efficacy. We reviewed the FIH trials of 67 anti-tumor products approved by the Food and Drug Administration between 2018 and 2023 and found that the "3 + 3" design remains the predominant dose-escalation method (66.2%). The number of patients receiving sub-therapeutic doses is positively correlated with the maximum tolerated dose (MTD) or maximum dose (MD) to starting dose ratio (P = 0.056) and the number of dose levels in trials (P < 0.001). In addition, the proportion of products with a high ratio in antibody drugs is higher than that in small molecules (P < 0.001). The MTD or MD exceeded the label dose by three or more doses in 22.03% of the products. In conclusion, optimizing the starting dose selection method, refining the way of determining doses, and finding alternative indicators to replace toxicity as the endpoints will increase the effectiveness and broaden the beneficiary scope.

Abstract OT-09-10: A phase I/II open-label, first-in-human, multicenter, dose escalation and dose expansion study of OP-1250 monotherapy in adult subjects with advanced and/or metastatic hormone receptor (HR)-positive, HER2-negative breast cancer

Abstract Background: Endocrine therapy has been the primary treatment modality for HR+, HER2- metastatic breast cancer (MBC). Endocrine agents are administered sequentially, either in combination with targeted therapy or as monotherapy. The majority of patients with HR+, HER2- MBC will develop endocrine resistant disease. More effective and less toxic therapies are needed for the treatment of endocrine resistant disease. OP-1250 is a small molecule Complete Estrogen Receptor ANtagonist (CERAN) that completely inactivates Estrogen Receptor (ER), blocks ER-driven transcriptional activity, inhibits ER-driven breast cancer cell growth, and induces degradation of ER. OP-1250 demonstrates anti-cancer activity in vitro and in vivo, including activity against metastases in the brain and in tumors with activating mutations in ESR1. OP-1250 is orally bioavailable with a favorable pharmacokinetic profile supportive of once-daily dosing. OP-1250 is hypothesized to completely antagonize ER resulting in superior efficacy compared to agents that only have partial antagonism of ER. Its favorable pharmacologic profile makes it an attractive agent for chronic use in patients with MBC. Trial design: This is a Phase I/II open-label, first-in-human study to determine the Dose Limiting Toxicity (DLT), Maximum Tolerated Dose (MTD) and/or Recommended Phase II Dose (RP2D), to characterize the safety and pharmacokinetic (PK) profile, and to determine the preliminary efficacy of OP-1250 in adult subjects with HR+, HER2- MBC. Treatment will consist of oral, once a day dosing and subject evaluation will be performed in 28-day cycles. This study comprises 2 parts. Part 1 (Dose Escalation) will evaluate the safety and pharmacology of a range of doses of OP-1250 administered orally to subjects and to determine the maximum tolerated dose (if any) and/or the RP2D. Cohorts of 3 to 6 subjects will be sequentially enrolled and monitored for DLTs during the first cycle of study treatment. Part 2 (Dose Expansion) will evaluate the preliminary activity of OP-1250. Patients with and without central nervous system (CNS) disease will be enrolled at the RP2D. This is designed using a Simon 2 Stage Design. Total accrual will be determined by the number of dose levels needed to identify the RP2D. Eligibility criteria:Males and females, age 18 or older, with ER+, HER2− advanced or MBCPrior treatment with endocrine therapyECOG performance status of 0 or 1For dose expansion the subject must have measurable disease according to Response Evaluation Criteria in Solid Tumors Criteria (RECIST) 1.1 ObjectivesPart 1 (Dose Escalation)To identify the DLT, MTD and/or RP2D of OP-1250To assess the safety and tolerability of OP-1250To assess the pharmacokinetics of OP-1250 Part 2 (Phase II: Monotherapy Expansion)Objective response rate (ORR) of OP-1250 in subjects with HR+, HER2- MBC who have progressed on endocrine therapy and have no evidence of central nervous system (CNS) metastases.To conduct a preliminary assessment of the antitumor activity (ORR) of OP-1250 in subjects with HR+, HER2- MBC who have progressed on endocrine therapy and have CNS disease. Assessment of response will be determined according to RECIST 1.1 and Response Assessment in Neuro-Oncology Brain Metastases (RANO-BM) criteria Correlative ScienceTo determine biomarker expression, such as, ER, PR, Ki67 and others in the most recently obtained archival tumor tissue sampleTo evaluate whether ESR1 in circulating tumor DNA (ctDNA) can be correlated with response and/or activity of OP-1250To examine ctDNA pre- and post-therapy for mutESR1 and PIK3CA variants, and other relevant markers For more information, please contact clinical@olemapharma.com Citation Format: Erika Hamilton, Carlos Alemany, Nancy U Lin, Pamela M Klein, Trinh Le, Peter J Kushner, Cyrus Harmon, Jo Anne Zujewski, Manish Patel. A phase I/II open-label, first-in-human, multicenter, dose escalation and dose expansion study of OP-1250 monotherapy in adult subjects with advanced and/or metastatic hormone receptor (HR)-positive, HER2-negative breast cancer [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr OT-09-10.

Evaluating the effects of design parameters on the performances of phase I trial designs.

Numerous designs have been proposed for phase I clinical trials. Although studies have compared their performances, few have considered the effects of changing design parameters. In this article, we review a few popular designs, including the 3 + 3, continuous reassessment method (CRM), Bayesian optimal interval (BOIN) design, and Keyboard design, and evaluate how varying design parameters (such as number of dose levels, target toxicity rate, maximum sample size, and cohort size) could impact the performances of each design through simulations. Excluded from our analysis is the mTPI-2 design, which operates in the same way as the Keyboard. Our results suggest that regardless of the choices of design parameters, the 3 + 3 design performs worse than the other ones, and BOIN and Keyboard have comparable performance to CRM. For any design, the performance varies with the choice of parameters. In particular, it improves as sample sizes increase, but the magnitude of benefit from increasing sample sizes varies substantially across scenarios. The impact of cohort size on design performances seems to have no clear direction. Therefore, BOIN and Keyboard designs are generally recommended due to their simplicity and good performance. With regard to choices of sample size and cohort size in designing a trial, it is recommend that simulations be performed for the particular clinical settings to aid decision making.

Reduced and optimized trial designs for drugs described by a target mediated drug disposition model

Monoclonal antibodies against soluble targets are often rich and include the sampling of multiple analytes over a lengthy period of time. Predictive models built on data obtained in such studies can be useful in all drug development phases. If adequate model predictions can be maintained with a reduced design (e.g. fewer samples or shorter duration) the use of such designs may be advocated. The effect of reducing and optimizing a rich design based on a published study for Omalizumab (OMA) was evaluated as an example. OMA pharmacokinetics were characterized using a target-mediated drug disposition model considering the binding of OMA to free IgE and the subsequent formation of an OMA–IgE complex. The performance of the reduced and optimized designs was evaluated with respect to: efficiency, parameter uncertainty and predictions of free target. It was possible to reduce the number of samples in the study by 30% while still maintaining an efficiency of almost 90%. A reduction in sampling duration by two-thirds resulted in an efficiency of 75%. Omission of any analyte measurement or a reduction of the number of dose levels was detrimental to the efficiency of the designs (efficiency ≤ 51%). However, other metrics were, in some cases, relatively unaffected, showing that multiple metrics may be needed to obtain balanced assessments of design performance.

Abstract OT1-4-02: Phase I study of the combination of BKM120 and olaparib for the treatment of high grade serous ovarian cancer (HGSC) and triple negative breast cancer (TNBC)

Abstract Background: In vivo synergy of the PI3-kinase inhibitor BKM120 with the PARP inhibitor olaparib has been observed using a mouse model of BRCA1-related breast cancer and sporadic TNBC (Juvekar et al and Ibrahim et al, Cancer Discovery 2012). In addition, olaparib has single agent activity in both HGSC and BRCA-associated breast cancer. The PI3kinase pathway is activated in both TNBC and HGSC (www.cancergenome.nih.gov). These preclinical and clinical data have served as the rationale for this phase I, multi-center study (NCT01623349) combining the oral PARP inhibitor olaparib with the oral PI3-kinase inhibitor BKM120 in patients with recurrent HGSC or recurrent TNBC. This study is being conducted through the Stand Up to Cancer (SU2C)'s Targeting PI3-kinase in Women's Cancers Dream Team. Study Design: This study has a 3 + 3 design, escalating if 0/3 or 1/6 participants have a dose limiting toxicity (DLT) during the first cycle of therapy (first 28 days). The study objectives are to determine the recommended phase II dose (RP2D) of daily continuous oral olaparib (using the tablet formulation) and BKM120, assess toxicities, safety, and preliminary activity of this combination, and determine pharmacokinetic profiles of both agents. In addition, there are several translational endpoints including elucidation of downstream signaling effects of the PI3-kinase pathway, examination of BRCA1 immunostaining, and assessment of BRCA1 promoter hypermethylation and somatic mutations in BRCA1 and BRCA2 using archived formalin fixed paraffin embedded (FFPE) tissue. Serial IL-8 and circulating DNA levels are also being monitored as well. Eligibility includes a diagnosis of recurrent TNBC or HGSC, PS 0 or 1, measurable or evaluable cancer, and normal lab values and organ function. Prior PARP inhibitor exposure is allowed. In addition, breast cancer or ovarian cancer patients with any histologic subtype are eligible if they have a known germline BRCA1 or BRCA2 mutation. At the RP2D, 10 pts each with a diagnosis of TNBC or HGSC will be enrolled to further determine safety and efficacy profiles in addition to more thoroughly studying translational endpoints. As of June 7, 2013, 16 patients have been enrolled into this study with a planned accrual of approximately 50 patients which may change based on number of dose levels tested during dose escalation. In addition, an amendment is pending that will add a second cohort studying the combination of olaparib and BYL719 based on robust pre-clinical activity observed in murine models which will increase our total accrual. Once this new cohort is open, both arms will enroll simultaneously. For further information, contact Ursula Matulonis at: umatulonis@partners.org. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr OT1-4-02.

A Review of Study Designs and Outcomes of Phase I Clinical Studies of Nanoparticle Agents Compared with Small-Molecule Anticancer Agents

Nanoparticles or carrier-mediated agents have been designed to prolong drug circulation time, increase tumor delivery, and improve therapeutic index compared to their small-molecule counterparts. The starting dose for phase I studies of small molecules and nanoparticles anticancer agents is based on the toxicity profile of the most sensitive species (e.g., rat or canine), but the optimal animal model for these studies of nanoparticles is unclear. The objective of this study was to evaluate the design, progression, and outcomes of phase I studies of nanoparticles compared with small-molecule anticancer agents. In preclinical studies, the maximum tolerated dose (MTD) in rats and dogs was evaluated for nanoparticles and their respective small molecules. In phase I clinical trials in patients with advanced solid tumors, the basis for starting dose, the number of dose escalations, number of patients enrolled, and the ratio of MTD to starting dose was determined for nanoparticles and small molecules. The mean ratio of MTD to starting dose in clinical phase I studies was significantly greater for nanoparticles (13.9 ± 10.8) compared with small molecules (2.1 ± 1.1; P = 0.005). The number of dose levels in a clinical phase I study was also significantly greater for nanoparticles (7.3 ± 2.9) compared with small molecules (4.1 ± 1.5; P = 0.008). The degree of dose escalation from starting dose to MTD was significantly greater for nanoparticles as compared with small-molecule anticancer drugs. These findings necessitate the need to identify the most appropriate preclinical animal model to use when evaluating nanoparticles toxicity.

A pool-adjacent-violators-algorithm approach to detect infinite parameter estimates in one-regressor dose–response models with asymptotes

Binary response models are often applied in dose–response settings where the number of dose levels is limited. Commonly, one can find cases where the maximum likelihood estimation process for these models produces infinite values for at least one of the parameters, often corresponding to the ‘separated data’ issue. Algorithms for detecting such data have been proposed, but are usually incorporated directly into in the parameter estimation. Additionally, they do not consider the use of asymptotes in the model formulation. In order to study this phenomenon in greater detail, we define the class of specifiably degenerate functions where this can occur (including the popular logistic and Weibull models) that allows for asymptotes in the dose–response specification. We demonstrate for this class that the well-known pool-adjacent-violators algorithm can efficiently pre-screen for non-estimable data. A simulation study demonstrates the frequency with which this problem can occur for various response models and conditions.

The influence of study design and sex-differences on results from developmental neurotoxicity studies of bisphenol A, implications for toxicity testing

Developmental neurotoxicity (DNT) of bisphenol A (BPA) has been investigated in a large number of studies. However, there are discrepancies in the results reported between the studies. The aim of this study was to identify and analyze factors that may contribute to these differences and to assess whether there are sex-differences in the sensitivity of certain endpoints or tests used in DNT-studies.Forty-four DNT studies of BPA were identified from the open literature. Details about study design and results from each study, as well as the criteria for DNT testing according to the standardized OECD test guideline (TG) 426, were collected in a database. This enabled systematic and detailed comparisons between studies as well as to the criteria and recommendations stated in TG 426. Multivariate analyses were also used to investigate how different factors of the study design contributed to differences in study results.The analyses showed behavioral effects were often observed for endpoints that are not required according to OECD TG 426, such as anxiety-related, social and sexual behaviors, especially at very low doses and in female offspring. On the other hand relatively few studies observed any effects on motor activity, which is commonly used in screening for neurotoxic effects in regulatory testing. However, varied and to some extent seemingly contradictory results have been reported in these studies, especially for endpoints related to motor activity and anxiety and exploration. Many studies were also poorly reported, limiting these analyses. No strong conclusions could be drawn from the multivariate analyses. A few factors of study design, such as the size of the dose and number of dose levels used and the use of litter or individual pup as statistical unit seemed to have some influence on study results.In conclusion, this analysis suggests that DNT-studies conducted according to the standardized OECD TG 426 may overlook sensitive effects of BPA, and possibly other potential endocrine disruptors, especially in female offspring.

Lack of compliance of European Public Assessment Reports to guidelines for paediatric drug development before the introduction of paediatric investigation plans

According to the International Conference on Harmonisation (ICH) and Food and Drug Administration (FDA) guidelines for paediatric clinical trials, bridging procedures can be used if disease progression, exposure-response relationships, and clinical endpoints are similar in adults and children. In these circumstances, confirmatory efficacy trials are not necessary; the evaluation of pharmacokinetics and safety ought to be sufficient for drug approval. The aim of this study was to assess whether the clinical trials and strategy for market approval authorisation (MAAs) in paediatric indications reflect the guidelines for bridging of adult data. A total of 95 European Public Assessment Reports (EPARs) published between 1995 and 2007 were reviewed. From every report, data extraction was performed according to the phase of development, scope of analysis, number of dose levels, dosage form, and demographics of the subjects enrolled in the trial. Data analysis consisted of an initial grouping of the studies by the degree of compliance to bridging guidelines. Our analysis reveals that only 66% of the trials (n = 174) can be classified as needed, while 22% of the trials (n = 59) could have been designed and performed differently from the approved protocol (partially required). Moreover, 12% (n = 30) of the studies were deemed completely unnecessary. A potential limitation in our study was that the dates of start and completion of the clinical studies were not available. Therefore, some EPARs have been included that may reflect common practice in the period that precedes the introduction of the ICH E11 guidelines. Yet, this should not obscure the points identified with regard to the lack of compliance to guidelines before the introduction of the paediatric legislation and the requirements for a paediatric investigation plan. Paediatric trials are desirable and necessary to address important unmet medical needs. However, the types of studies supporting regulatory approval do not always reflect the recommendations available in paediatric guidelines, which allow for extrapolation and bridging approaches. This situation may be explained by the lack of awareness about the prerequisites for the use of bridging concepts and of a clear process for evaluating different strategies in paediatric development.

Abstract 371: Comparison of toxicity and study design issues of nanoparticle and small molecule anticancer agents in preclinical models and phase I clinical trials

Abstract Background: Carrier-mediated agents (CMA) are classified as nanoparticles, nanosomes (nanoparticle sized liposomes) and conjugates. Anticancer CMA offer many unique advantages over their traditional small molecule (SM) counterparts, including improved solubility, longer duration of exposure, tumor-selective delivery, increased antitumor response and reduced toxicity. The interpatient variability in pharmacokinetics (PK) and pharmacodynamics (PD) associated with nanoparticles are greater than that observed with SM. The starting dose for phase I studies of SM and CMA anticancer agents is based on the toxicity profile of the most sensitive species (e.g. rat or canine). However, the optimal animal model for toxicologic and pharmacologic studies of CMA is unclear. Our preliminary data suggests that dogs are highly sensitive to nanoparticles and thus may not be an appropriate animal model in determining the starting dose of CMA for phase I clinical trials. We evaluated preclinical toxicology data and how this affected the design and progression of phase I studies of CMA compared to SM anticancer agents. Methods: In preclinical studies, the maximum tolerated dose (MTD) in rats and dogs was evaluated for CMA and their respective SM. In phase I clinical trials of CMA and their respective SM in patients with advanced solid tumors, the basis for starting dose, the number of dose escalations, number of patients and the ratio of MTD to starting dose was determined. Results: Starting dose in phase I studies of CMA was based on dogs. Conclusions: The degree of dose escalation from starting dose to MTD was significantly greater for CMA compared to SM drugs. This was also associated with a significantly greater number of dose levels, patients, and time required to complete phase I studies of CMA compared to SM. These findings necessitate the need to identify the most appropriate preclinical animal model to use to evaluate CMA toxicity in phase I studies. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 371. doi:10.1158/1538-7445.AM2011-371

Bias Reduction in Logistic Dose-Response Models

In generalized linear models, such as the logistic regression model, maximum likelihood estimators are well known to be biased at smaller sample sizes. When the number of dose levels or replications per dose is small, bias in the maximum likelihood estimates can lead to very misleading results and the model often fails to converge. In order to correct the bias present in the maximum likelihood estimates and the problem of nonconvergence, the penalized maximum likelihood estimator is considered. Simulations compare the fit and empirical confidence levels of inferences made from the maximum likelihood and penalized maximum likelihood based models.

Abstract A98: Safety, efficacy and efficiency of phase 1 single agent trials using the accelerated titration (ATD) versus modified Fibonacci designs (STD)

Abstract Background: Phase 1 trials of oncology drugs should balance safety, ethics and efficiency; however there is limited consensus on preferred designs for first-in-human studies (FIH). Methods: We conducted a review of the major oncology journals to identify phase 1 oncology trials using ATD and the standard modified Fibonacci design (STD) published between 2000–08. Trials were limited to FIH, single agent, and patients with solid tumors, and used animal toxicology studies to determine starting dose. Data on agent, class, design, starting dose rationale, dose levels, toxicities and responses were abstracted and compared to determine the efficiency, safety, and response rate of the ATD and STD. Results: We identified 46 ATD (1270 patients) and 90 SDT (3035 patients) trials published between 2000–8. Fewer patients were treated at low dose levels (mean difference = 6) and overall (mean n = 28 and 34), there was no significant increase in the number of patients treated above the recommended phase 2 dose (RP2D) or in the mean number of treatment related-deaths on trials using the ATD versus STD (1.3% vs. 1%). Between ATD and STD, the mean number of dose levels evaluated was the same; however, the mean fold dose range evaluated was greater with the ATD. The objective response rate (ORR) was significantly higher among evaluable patients on ATD versus STD trials (2.7% vs. 1.7%, p=.02 1-sided). These results suggest a gain in efficiency both by limiting numbers of patients and numbers of dose levels required to establish the RP2D and by improving the number of patients treated at biological active doses. Most ATD trials used designs 3 and 4, which utilize single patient cohorts and 100% dose escalations and 48% of the ATD trials utilized intra-patient dose escalation (IPDE); thus, more patients overall were treated closer to the RP2D. Both ORR and numbers of treatment-deaths were higher in trials using IPDE. Excess treatment-related deaths did not appear to occur for IPDE patients. Total death rates were 1.71% for STD trials and 1.42% for ATD trials (0.94% for trials without IPDE and 1.90% for trials with IPDE) suggesting either an increase in treatment- versus disease-related deaths in IPDE trials or potential bias in attribution. Proportionately more ATD versus STD trials used 1/10th the LD10 from murine models to determine the starting doses, tested agents that could be classified within traditional chemotherapy classes and reported dose-limiting hepatotoxicity and mucositis. Conclusions: Our results support the safety, efficiency and efficacy of the ATD designs and suggest that the ATD is preferable to the STD. The role of IPDE requires further evaluation. ATD versus STD Trials ATD STD 1-s. p-value Total Trials (n) 46 (22 with IPDE) 90 Total patients (n) 1270 3035 Mean patients/trial 27.6 33.7 p=.01 Overall response rate among evaluable pts [restricted to IPDE trials in ()] 2.73% 1.69% p=.02 (3.49%) (p=.005) Total treatment-related deaths 17 [1.34%] 30 p=.20 [IPDE trials in ()] (11 [1.75%]) [.99%] (p=.08) Mean dose levels/trial 7.8 7.3 Median dose fold-increase/trial 16.8 12.0 Mean pts &amp;gt; recommended dose 8.2 7.2 Mean doses &amp;gt; recommended dose 1.5 1.3 Citation Information: Mol Cancer Ther 2009;8(12 Suppl):A98.

The benchmark dose approach in food risk assessment: Is it applicable and worthwhile?

The benchmark dose (BMD) approach is being increasingly used in the area of food risk assessment because it offers several advantages compared to the conventional no-observed-adverse-effect-level approach. The aim of this work was to check the applicability of the BMD approach on toxicity data available from pesticides, mycotoxins and natural toxins. Based on toxicological evaluations, the pivotal study was identified. Detailed data from the original study were retrieved and used for BMD modelling. Twenty-five studies used for BMD modelling were analysed with regard to study design: total number of animals, number of dose levels, and spacing between dose levels. The quality of the modelled endpoints was evaluated according to the following aspects: BMD/BMDL ratio, test for goodness of fit and BMD in the range of dose levels. If one of these aspects was not fulfilled, the BMD derived from this endpoint was considered to be uncertain to some extent and corresponding modelled data sets were examined. The present work demonstrates that the BMD approach is in principle applicable to pesticides, mycotoxins, and natural toxins. Although large differences relating to data availability and data quality were noticed, 69 of 82 modelled endpoints (84%) fulfilled the three quality aspects of BMD modelling.

Range and trend of expected toxicity level (ETL) in standard A + B designs: A report from the children's oncology group

The traditional algorithm-based 3 + 3 designs are most widely used for their practical simplicity in phase I clinical trials. At early stage, a common belief was that the expected toxicity level (ETL) at the maximum tolerated dose (MTD) should be 33% [Storer, B. Design and analysis of phase I clinical trials. Biometrics 1989;45;925–937, Gorden, N., Willson, J. Using toxicity grades in the design and analysis of cancer phase I clinical trials. Statistics in Medicine 1992; 11: 2063–2075, Mick, R. Phase I Clinical Trial Design. In Schilsky, R., Milano, G., Ratain, M., eds. Principles of Antineoplastic Drug Development and Pharmacology New York, NY: Marcel Dekker, 1996; 29–36]. Recently, Kang and Ahn [Kang, S., Ahn, C. The expected toxicity rate at the maximum tolerated dose in the standard phase I cancer clinical trial design. Drug Information Journal 2001; 35:1189–1199, Kang, S., Ahn, C. An investigation of the traditional algorithm-based designs for phase I cancer clinical trials. Drug Information Journal 2002; 36:865–873] found that the ETL is between 17% and 21% and He et al [He, W., Liu, J., Binkowitz, B., Quan, H. A model-based approach in the estimation of the maximum tolerated dose in phase I cancer clinical trials. Statistics in Medicine 2006; 25(12):2027–42] further reported that the ETL ranges from 19% to 24%. However they only investigated designs where the number of dose levels was at most 20. It has practical significance in designing and conducting phase I clinical trial to definitely assess the full range and trend of ETL by all possible number of tested dose levels in traditional algorithm-based A + B designs, especially 3 + 3 designs. In this simulation study, we originally find that the ETL decreases monotonically from about 30% to 0% as the number of dose levels increase from 3 to infinity, which will correct the inaccuracy in the common belief among phase I trial investigators. To help better design and conduct phase I trials, we create a table as a reference for the association between ETL and number of dose levels considered in a design when the exact shape of the dose-toxicity relationship is not well understood. We conclude that the number of specified dose levels is an important factor affecting substantially the ETL at MTD and recommend that fewer than 20 dose levels be designated.

Is the Continual Reassessment Method superior to the standard 3+3 dose escalation scheme?

2559 Background: Bayesian phase I designs have received increased attention. The objective is to determine if the Continual Reassessment Method (CRM) is superior to the standard 3+3 dose escalation scheme (SM) in terms of sample size, precision and study duration. Methods: Comparisons were carried out through 1,000 simulated trials under different scenarios by varying the number of dose levels from 5 to 8 and the location of the true MTD. The hyperbolic tangent was assumed as the dose-toxicity curve and the target toxicity rate was 25%. Four Bayesian designs were evaluated: original CRM; CRM that is restricted in no more than one dose level increase; two-stage CRM that follows an arbitrary escalation plan at the beginning of the trial until a DLT is observed and then switches to CRM; and CRM in cohorts of three patients. CRM was assessed in two alternative schemes: with a fixed sample approach of 20 patients or with a stopping rule that halts the trial when a specified amount of precision around the estimated toxicity rate at the MTD (ETR) is reached. Standard endpoints were evaluated: accuracy, precision, safety, total sample size, and minimum study duration. Results are independent of the increment between dose levels. Results: SM found the true MTD on average 20% (range 8–28%) fewer times compared to CRM-based methods in almost all scenarios. All CRM-based methods treated more patients at the MTD compared to SM (42% vs 11%) except when the true MTD is at the second dose level (36 vs 35%). Number of DLTs was the smallest under the SM. CRM and SM are similar in terms of total sample size when testing 5 dose levels, although SM requires shorter duration. As the number of dose levels increases to 8, CRM completes the trial faster than SM when used with a fixed sample of 20 patients; except when the MTD is among the lower levels. Using the stopping rule approach with a maximum of 30 patients, 66–78% of trials have not met the required precision around the ETR. Conclusions: All CRM-based methods are an improvement over SM in terms of accuracy and optimal dose allocation, except when the true MTD is among the lower levels. When the number of dose levels is limited (eg 5), a sample of 20 is sufficient by both methods in order to find the MTD. However, a sample size of 20–30 is not sufficient to achieve a narrow confidence interval around the ETR. No significant financial relationships to disclose.

Choice of designs and doses for early phase trials.

This paper reviews state-of-art statistical designs for dose-escalation procedures in first-into-man studies. The main focus will be on studies in oncology, as most statistical procedures for phase I trials have been proposed in this context. Extensions to situations such as the observation of bivariate outcomes and healthy volunteer studies are also discussed. The number of dose levels and cohort sizes used in early phase trials are considered. Finally, this paper raises some practical issues for dose-escalation procedures.

Selection of number of dose levels and its robustness for binary response data

Müller & Schmitt (1990) have considered the question of how to choose the number of doses for estimating the median effective dose (ED50) when a probit dose-response curve is correctly assumed. However, they restricted their investigation to designs with doses symmetrical about the true ED50. In this paper, we investigate how the conclusions of Müller & Schmitt may change as the dose designs become slightly asymmetric about the true ED50. In addition, we address the question of the robustness of the number of doses chosen for an incorrectly assumed logistic model, when the dose designs are asymmetric about the assumed ED50. The underlying true dose-response curves considered here include the probit, cubic logistic and Aranda- Ordaz asymmetric models. The simulation results show that, for various underlying true dose-response curves and the uniform design density with doses spaced asymmetrically around the assumed ED50, the choice of as many doses as possible is almost optimal. This agrees with the results obtained for a correctly assumed probit or logistic dose-response curve when the dose designs are symmetric or slightly asymmetric about the ED50.

Comparison of available benchmark dose softwares and models using trichloroethylene as a model substance

By using trichloroethylene as a model substance the U.S. EPA benchmark dose software was compared to the software by Crump and the software by Kalliomaa. Dose–response and dose–effect data on the liver, kidneys, central nervous system (CNS), and tumours were selected for the evaluation. Based on the present study the U.S. EPA software is preferable to the other softwares for dichotomous data. A wider range in benchmark doses was often observed for dichotomous data when the numbers of dose levels were limited. The log-logistic model in most cases gave the best fit when ranking the dichotomous models. In addition, the log-logistic model often implied a more conservative benchmark dose. For continuous data it was more difficult to find a model describing the data. The softwares by Kalliomaa and by the U.S. EPA offered the best opportunities for benchmark dose modelling of continuous data. Flexible models, like the Hill- and the Mult model, are needed for S-shaped continuous data but these models demand more dose levels in order to describe the data. Since the number of dose levels are important for model selection study design is important and should be further evaluated.

Dose and litter allocations in the design of teratological studies for detecting hormesis.

Hormesis is being recognized in the field of toxicology due to the stimulating effects of some toxic compounds at low exposure levels. Therefore, it is desirable that experimental designs for toxicological studies be flexible enough to aid in the detection of hormetic effects. Current designs may still not have enough power to do this. A simulation study was conducted to determine teratological study designs that would yield more power over standard designs in detecting hormesis. Developmental toxicity endpoints of interest are the number of dead/resorbed or malformed fetuses in a litter. The simulation designs mimic teratological experiments in terms of sample size and number of dose levels. Modified designs with even dose spacing at low levels and reallocated litters are investigated to determine the power of hormetic detection. Designs with reallocated litters (with more litters at low exposure levels than at high levels) and even dose spacing have more power than those with equal litters per group and uneven dose spacing. Through appropriate modifications of current experimental designs, such as reallocation of litters and even dose spacing, we can better detect hormetic effects.

Design of Phase I and II Trials in Cancer

This paper provides a survey of current design strategies for Phase I and II trials of chemotherapeutic agents and biological response modifiers pertinent to cancer therapy. Definitions of Phase I and II trials are first given. A typical Phase I trial is described using an example of trial testing a new colony stimulating factor (CSF), which had the goal of defining toxicity with a limited number of dose levels. Two designs of Phase II trials are then described, first, the traditional design of Gehan, and then the multiple testing procedure of Elfring and Schultz and Fleming. The advantages and disadvantages of these designs are discussed. Some extensions are then suggested, such as a design for a Phase I–II trial and a randomized Phase II trial that may be extended into a Phase III trial if the treatments have sufficient efficacy. Early statistical input into Phase I and II trials of cancer treatment is needed to assure proper clinical trial design.