Dose Limiting Toxicity (DLT)

Some concepts about DLT in early phase dose finding studies.

The definition of a Dose Limiting Toxicity will be trial specific based on a predefined set of unacceptable AEs that are observed. The predefined list will specify if events are possibly study treatment-related or if events regardless of study treatment causality classify as a DLT.

Note: The DTL definition and what constitutes a DLT are included as a section in the protocol.

Toxicities are graded according to the most current (at the time of protocol development) National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events (CTCAE).

Dose Related Definitions

Maximum Tolerated Dose (MTD), Maximum Feasible Dose (MFD), Maximum Administered Dose (MAD) and Recommended Phase 2 Dose (RP2D) are not patient endpoints but are often associated with the primary objective of Phase 1 trials.

Maximum Tolerated Dose (MTD)

The general definition of MTD is the dose, d, for which the probability of a DLT as a function of dose, \(ψ(d)\), is equal to a specified target probability.

There are various ways to design Phase 1 studies to estimate the MTD.

For the rule-based 3+3 design, the MTD estimate is the highest dose associated with the occurrence of DLTs in <33% of patients. Typically, the MTD estimate is the dose level at which 0/6 or 1/6 patients experience a DLT during the first K treatment cycle(s) with the next higher dose having at least 2 of 3 to 6 patients experiencing DLTs. K is often chosen to be 1, but should be >1 if delayed or cumulative toxicities are anticipated. Note that the target probability is unspecified for the 3+3 design but implicitly is between 1/6 and 1/3.

Up and Down designs such as the Bayesian Logistic Regression Model (BLRM) and Continual Reassessment Method (CRM) and their modifications are common alternatives to the 3+3 design. These have specific target probabilities. Estimates from these methods are model based and have better properties than the 3+3 design. ^c,d,e

The modified toxicity probability interval (mTPI/mTPI-2) methods ^a,b which are model-assisted designs in most circumstances provide improvement over the 3+3 design with a lower risk of exposing patients to toxic doses above the MTD and a higher probability of identifying the correct MTD, and are easier to implement than CRM type of designs.

Note: the software EAST has the mTPI, BLRM and CRM designs as part of the Escalate module.

Most of the above designs have extensions for addressing combination of two or more drugs.

Maximum Feasible Dose (MFD)

Maximum Feasible Dose is the protocol specified highest dose that may be administered for practical reasons even if toxicity is not observed. If escalation continues to the MFD and the target toxicity rate has not been reached, there are no more escalations and the trial ends with the conclusion that the MTD is greater than or equal to the MFD.

Maximum Administered Dose (MAD)

Maximum Administered Dose is the highest dose actually administered in the trial for the study drug(s) being escalated.

Typically this is a dose above the MTD estimate, and may be the MFD.

Recommended Phase 2 Dose (RP2D)

The Recommended Phase 2 Dose is the dose chosen for further study based on Phase 1 results. The RP2D is a dose chosen that is at or below the MTD. Additional patients may be enrolled in the trial at the MTD after the MTD is reached to provide more precise estimates of PK or other biologic parameters or additional anti-tumor or safety information. In such cases the protocol should include a sample size justification for the additional patients enrolled.

Adverse Events (AEs) – Grades and Definition of Treatment-Emergence

Grades for specific AEs are defined according to CTCAE, published by the U.S. NCI. The criteria can be found on the CTEP website: http://ctep.info.nih.gov/reporting/ctc.html. The protocol should indicate which version of CTCAE is to be used.

The on-treatment period is defined as the period that starts with the first dose of study treatment and ends at min(last dose of study treatment + xx days, start of new anti-cancer therapy – 1 day). The definition of anti-cancer therapy and the “xx” days should be pre-specified in the protocol and/or SAP.

Two definitions of treatment‑emergent AEs (TEAEs) can be considered as described below. The definition that is used for a particular trial or drug program should be aligned with input from Health Authorities and the applicable internal standards.

1. TEAEs are AEs with onset date during the on-treatment period

2. TEAEs are AEs that occur for the first time during the on-treatment period or AEs that were observed prior to the start of study treatment but increased in severity during the on-treatment period.

Reference

[a] Ji Y, Liu P, Li Y and Bekele BN. A modified toxicity probability interval method for dose-finding trials. Clinical Trials 7: 653-663, 2010

[b] Ji Y, Wang, SJ. Modified toxicity probability interval design: a safer and more reliable method than the 3+3 design for practical phase I trials. Journal of Clinical Oncology 31(14): 1785-91, 2013.

[c] Goodman S, Zahuak, M and Piantadosi S. Some practical improvements in the continual reassesment method for phase I studies, Statistics in Medicine 14:1149-1161, 1995

[d] Storer B. Choosing a Phase I Design. In Handbook of Statistics in Clinical Oncology, Second Edition Eds. Crowley J, Ankerst D. CRC Press, Boca Raton, 2006

[e] O’Quigley J. Phase I and Phase I/II Dose Finding Algorithms Using Continual Reassessment Methods In Handbook of Statistics in Clinical Oncology, Second Edition Eds. Crowley J, Ankerst D. CRC Press, Boca Raton, 2006