Phase I: Overview and Recent Trial Designs

Drug development enterprise is struggling because of prohibitively high costs and slow progress. There is urgent need for adoption of novel adaptive designs to improve the efficiency and success of clinical trials. A major barrier is that many conventional designs are inadequate for modern drug development, yet most novel adaptive designs are difficult to understand, require complicated statistical modeling, demand complex computation, and need expensive infrastructure for implementation. The objective of this article is to introduce and review a class of novel adaptive designs, known as model-assisted designs, to remove this barrier and increase the use of novel adaptive designs. Model-assisted designs enjoy superior performance comparable to more complicated, model-based adaptive designs, but their decision rule can be pretabulated and included in the protocol-thus implemented as simply as the conventional designs. We review state-of-the-art model-assisted designs for phase I clinical trials for single-agent, drug-combination and late-onset toxicity scenarios. We also briefly introduce model-assisted designs for phase II trials to handle binary, coprimary endpoints and delayed response. Freely available user-friendly software and trial examples (trialdesign.org) facilitate the adoption of model-assisted designs.

Recently, the US Food and Drug Administration Oncology Center of Excellence initiated Project Optimus to reform the dose optimization and dose selection paradigm in oncology drug development. The agency pointed out that the current paradigm for dose selection-based on the maximum tolerated dose (MTD)-is not sufficient for molecularly targeted therapies and immunotherapies, for which efficacy may not increase after the dose reaches a certain level. In these cases, it is more appropriate to identify the optimal biological dose (OBD) that optimizes the risk-benefit tradeoff of the drug. Project Optimus has spurred tremendous interest and urgent need for guidance on designing dose optimization trials. In this article, we review several representative dose optimization designs, including model-based and model-assisted designs, and compare their operating characteristics based on 10,000 randomly generated scenarios with various dose-toxicity and dose-efficacy curves and some fixed representative scenarios. The results show that, compared with model-based designs, model-assisted methods have advantages of easy-to-implement, robustness, and high accuracy to identify OBD. Some guidance is provided to help biostatisticians and clinicians to choose appropriate dose optimization methods in practice.

The paradigm of early phase dose-finding trials has evolved in recent years. Innovative dose-finding designs and protocols which combine phases I and II are becoming more popular in health research. However, the quality of these trial protocols is unknown due to a lack of specific reporting guidelines. Here, we evaluated the reporting quality of dose-finding trial protocols. We conducted a cross-sectional study of oncology and non-oncology early phase dose-finding trial protocols posted on ClinicalTrials.gov in 2017-2023. A checklist of items comprising: 1) the original 33-items from the SPIRIT 2013 Statement and 2) additional items unique to dose-finding trials were used to assess reporting quality. The primary endpoint was the overall proportion of adequately reported items. This study was registered with PROSPERO (no: CRD42022314572). A total of 106 trial protocols were included in the study with the rule-based 3+3 being the most used trial design (39.6%). Eleven model-based and model-assisted designs were identified in oncology trials only (11/58, 19.0%). The overall proportion of adequately reported items was 65.1% (95%CI: 63.9-66.3%). However, the reporting quality of each individual item varied substantially (range 9.4%-100%). Oncology study protocols showed lower reporting quality than non-oncology. In the multivariable analysis, trials with larger sample sizes and industry funding were associated with higher proportions of adequately reported items (all p-values <0.05). The overall reporting quality of early phase dose-finding trial protocols is suboptimal (65.1%). There is a need for improved completeness and transparency in early phase dose-finding trial protocols to facilitate rigorous trial conduct, reproducibility and external review. None.

Historically, oncology dose selection was driven by the identification of the maximum tolerated dose, often resulting in significant toxicity. The advent of molecular targeted therapies and immunotherapies has created opportunities to achieve similar efficacy at a lower, more tolerable dose. Recent FDA draft guidance highlights the importance of dose optimization in drug development. Building on this, we present a novel framework categorizing drugs into four molecular classes based on their mechanisms of action: (1) small molecule targeted therapies and antibody-drug conjugates, (2) large molecule antagonists, (3) cancer immunotherapy agonists, and (4) molecules with limited or no single-agent activity. Unique considerations and strategies for dose optimization are detailed for each class, supported by case examples. We propose integrating proof of activity as a gating criterion for dose expansion to inform robust dose decisions and maximize the use of patient resources. This perspective underscores the integration of preclinical and clinical data—including pharmacokinetics, pharmacodynamics, and patient-reported outcomes—combined with quantitative pharmacology and statistical modeling, to optimize doses effectively while balancing efficacy and safety. While no one-size-fits-all approach exists, this tailored guidance for specific mechanistic classes provides balanced approaches to dose optimization that ultimately enhance the probability of success in drug development. Citation Format: Neekesh Dharia, Jiawen Zhu, Amy Schroeder, Sabine Frank, Christophe Boetsch, Candice Jamois, Nastya Kassir, Koorosh Korfi, Elizabeth Punnoose, Anjali Vaze, Peter Trask, Pritti Gosai, Jane Fridlyand, Chunze Li. Refining oncology dose selection: A framework for tailored dose optimization in the era of targeted and immunotherapies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2514.

The sample sizes of phase I trials are typically small; some designs may lead to inaccurate estimation of the maximum tolerated dose (MTD). The objective of this study was to propose a metric assessing whether the MTD decision is sensitive to enrolling a few additional subjects in a phase I dose-finding trial. Numerous model-based and model-assisted designs have been proposed to improve the efficiency and accuracy of finding the MTD. The Fragility Index (FI) is a widely used metric quantifying the statistical robustness of randomized controlled trials by estimating the number of events needed to change a statistically significant result to non-significant (or vice versa). We propose a modified Fragility Index (mFI), defined as the minimum number of additional participants required to potentially change the estimated MTD, to supplement existing designs identifying fragile phase I trial results. Three oncology trials were used to illustrate how to evaluate the fragility of phase I trials using mFI. The results showed that two of the trials were not sensitive to additional subjects' participation while the third trial was quite fragile to one or two additional subjects. The mFI can be a useful metric assessing the fragility of phase I trials and facilitating robust identification of MTD.

Strategies for Dose Optimization: Views From Health Care Systems.

Despite the mounting evidence for the superior performance of model-based dose-finding designs in identifying the maximum tolerated dose compared to traditional up-and-down designs, its use is still very limited in medical literature as there remain several challenges in implementation. We introduce the use of the Dose Transition Pathways, and demonstrate how it can be used to overcome several commonly faced challenges, including: Firstly, the challenge of communicating to clinicians about how the design works and the rationale for using a method with substantially higher level of complexity. Secondly, operational challenges including trialists' impressions that dose recommendations come from a “black-box”, which contrast unfavourably with the transparent simple rules of a traditional 3+3 design. Possible delays may thus result, particularly for each dose recommendation. Thirdly, methodological challenges include how to choose an appropriate bespoke adaptive design that is applicable in practice. We will illustrate how the Dose-Transition Pathway can be used to aid in planning and designing an adaptive model-based dose-finding trial, specifically tailored to its requirements, with a case study of a Phase I Acute Myeloid Leukaemia and Myelodysplastic Syndrome trial. By aiding trialist's understanding of the working of such adaptive designs in practice, as well as achieving a seamless, clear transition for each dose-recommendation, the Dose Transition Pathways will serve as a valuable tool in the implementation of adaptive model-based designs in dose-finding trials.

The Anti-BCMA Antibody-Drug Conjugate Hdp-101 with a Novel Amanitin Payload Shows Promising Initial First in Human Results in Relapsed Multiple Myeloma

&lt;div&gt;Abstract&lt;p&gt;Cytokine release syndrome (CRS) is a common acute toxicity in T-cell therapies, including T-cell–engaging bispecific antibodies (T-BiSp). Effective CRS management and prevention are crucial in T-BiSp development. Required hospitalization for seven of the nine approved T-BiSp and the need for clinical intervention in severe cases highlight the importance of mitigation strategies to reduce health care burden and improve patient outcomes. In this review, we discuss the emerging evidence on CRS mitigation, management, and prediction. We cover different strategies for dose optimization, current and emerging (pre) treatment strategies, quantitative pharmacology tools used during drug development, and biomarkers and predictive factors. Insights are gleaned on step-up dosing and formulation effects on CRS and CRS relationships with cytokine dynamics and drug levels gathered through a review of T-BiSp licensing applications and emerging data from conferences and publications.&lt;/p&gt;&lt;/div&gt;

1. Infections and related sepsis are two of the most prevalent issues in the care of critically ill patients, with mortality as high as 70%. Appropriate antibiotic selection, as well as adequate dosing, is important to improve the clinical outcome for these patients. 2. β-Lactams are the most common antibiotic class used in critically ill sepsis patients because of their broad spectrum of activity and high tolerability. β-Lactams exhibit time-dependent antibacterial activity. Therefore, concentrations need to be maintained above the minimum inhibitory concentration (MIC) of pathogenic bacteria. β-Lactams are hydrophilic antibiotics with small distribution volumes similar to extracellular water and are predominantly excreted through the renal system. 3. Critically ill patients experience a myriad of physiological changes that result in changes in the pharmacokinetics (PK) of hydrophilic drugs such as β-lactams. A different approach to dosing with β-lactams may increase the likelihood of positive outcomes considering the pharmacodynamics (PD) of β-lactams, as well as the changes in PK in critically ill patients. 4. The present review describes the strategies for dose optimization of β-lactams in critically ill patients in line with the PK and PD of these drugs.

Introduction: CT imaging is essential in oncology, yet its widespread use raises concerns about radiation exposure. Therefore, dose optimization strategies are needed without compromising image quality. Recent advances in iterative reconstruction and AI-based algorithms offer promising solutions. Methods: This retrospective study was conducted between October 2023 and June 2024 at IRCCS CROB Rionero in Vulture. Examinations were performed on an older-generation Toshiba scanner and two new-generation Canon Aquilion ONE Genesis 320-slice scanners. The study included 14 brain CTs, 28 chest CTs, and 15 triphasic chest-abdomen-pelvis CTs, totaling 174 scans. Dose indicators (CTDIvol and DLP) were recorded and compared between the two CT systems and against diagnostic reference levels (DRLs) available in the literature. Results: The implementation of AI-based reconstruction techniques led to significant dose reactions. In chest and chest-abdomen-pelvis examinations, CTDIvol decreased by up to 82% and DLP by 70% compared to the older system. For brain imaging, iterative reconstruction was preferred to avoid excessive noise suppression associated with AI approaches, resulting in optimized protocols that preserved diagnostic image quality while enhancing anatomical detail. Conclusion: Integrating AI algorithms and iterative reconstruction techniques in CT imaging can substantially reduce radiation exposure while maintaining excellent diagnostic reliability.

Phase I trial designs in oncology aim to determine the maximum tolerated doses for agents of interest. Traditionally, dose-finding designs for phase I trials can be classified as rule-/algorithm- or model-based designs. In Chap. 2, we focused on rule-based designs; in the present chapter, we focus on model-based designs. Model-based designs assume a certain statistical model for the monotonic dose–toxicity relationship to borrow strength across doses, and originate from the continual reassessment method (CRM). The present chapter provides a detailed description of the concepts, theories, properties, and advantages and disadvantages of the CRM, and also overview designs that are related to the CRM or that constitute extended versions of this method.

This study compared the association between weighted computed tomography dose index (CTDIw) and effective milliampere-seconds (effective mAs) with the effective dose (ED) received by patients during thoracic computed tomography (CT) examinations. Retrospective data from breast cancer patients were obtained from The Cancer Imaging Archive (TCIA) and processed using IndoseCT software to extract CTDIw, effective mAs, and ED values. Descriptive statistics, scatter plots, and simple linear regression analyses were performed to evaluate the relationships between these parameters. The correlation between effective mAs and CTDIw was moderately strong (R² = 0.6774), consistent with the principle that higher tube current–time settings generally increase scanner-reported dose indices. CTDIw and ED showed a very strong correlation (R² = 0.978), reflecting the close link between standardized dose metrics and estimated patient dose. The correlation between effective mAs and ED (R² = 0.7505) was stronger than that between effective mAs and CTDIw, but still lower than the CTDIw–ED correlation. These results indicate that, in this dataset, CTDIw was a more consistent predictor of ED than effective mAs. While both parameters are relevant for dose assessment, CTDIw may be a more reliable reference for estimating ED in similar thoracic CT protocols. Although this analysis was limited to retrospective data from a specific patient group, the findings provide practical insights that can support protocol evaluation and dose optimization strategies. Further studies involving larger and more diverse datasets would be beneficial to confirm and extend these observations.

Despite more than two decades of publications that offer more innovative model-based designs, the classical 3 + 3 design remains the most dominant phase I trial design in practice. In this article, we introduce a new trial design, the Bayesian optimal interval (BOIN) design. The BOIN design is easy to implement in a way similar to the 3 + 3 design, but is more flexible for choosing the target toxicity rate and cohort size and yields a substantially better performance that is comparable with that of more complex model-based designs. The BOIN design contains the 3 + 3 design and the accelerated titration design as special cases, thus linking it to established phase I approaches. A numerical study shows that the BOIN design generally outperforms the 3 + 3 design and the modified toxicity probability interval (mTPI) design. The BOIN design is more likely than the 3 + 3 design to correctly select the MTD and allocate more patients to the MTD. Compared with the mTPI design, the BOIN design has a substantially lower risk of overdosing patients and generally a higher probability of correctly selecting the MTD. User-friendly software is freely available to facilitate the application of the BOIN design. Clin Cancer Res; 22(17); 4291-301. ©2016 AACR.

Concerns have been recognized about the operating characteristics of the standard 3 + 3 dose-escalation design. Various innovative phase 1 trial designs have been proposed to address the issues and new challenges posed by molecularly targeted agents. However, in spite of these proposals, the conventional design is still the most widely utilized. A review of the literature of phase 1 trials and relevant statistical studies was performed. Beyond statistical simulations, sparse clinical data exist to support or refute many of the shortcomings ascribed to the 3 + 3 rule method. Data from phase 1 trials demonstrate that traditional designs identified the correct dose and relevant toxicities with an acceptable level of precision in some instances; however, no single escalation method was proven superior in all circumstances. Design selection should be guided by the principle of slow escalation in the face of toxicity and rapid dose increases in the setting of minimal or no adverse events. When the toxicity of a drug is uncertain or a narrow therapeutic window is suggested from preclinical testing, then a conservative 3 + 3 method is generally appropriate. However, if the therapeutic window is wide and the expected toxicity is low, then rapid escalation with a novel rule- or model-based design should be employed.