DLT: Clinical Trial Terminology

Dose-limiting toxicity (DLT) meaning

In clinical research, DLT is an abbreviation for dose-limiting toxicity, which refers to adverse reactions elicited by an investigational drug that are serious enough to prevent researchers from increasing its dosage for a clinical trial.¹ In other words, a DLT is an instance of toxicity that necessitates limiting the dose of the drug. This is an important concept in drug development and early stage clinical trials, and is vital for ensuring the safety of trial participants.

Dose escalation studies: Determining the maximum tolerable dose (MTD)

Typically, the first stage of clinical trials for a new medicine (i.e., a phase I trial) is when researchers study the tolerability and potential toxicity of the novel drug to identify the dosages that are safe.

Such trials often employ a method known as dose escalation, which involves incrementally increasing the dosage of the drug under study to find the highest dose that can be given to participants without causing unacceptable adverse reactions. The maximum dose identified through this method is called the maximum tolerable dose, or MTD, and is used to establish recommended dosages and dose schedules for subsequent phase II clinical trials.²

Dose escalation studies are closely monitored and must follow strict regulatory guidelines to ensure participant safety is prioritized. The goal of these studies is to identify safe dosages, and not to treat patients.

What is DLT in clinical trials?

When investigating a novel drug in humans for the first time (sometimes called “first-in-human” studies), researchers often have little information available for directly determining a safe and effective dose. A common way to identify this dose is to increase it slowly and monitor attentively for any DLT. Clinical trials, therefore, use DLTs as an indication that a given dose may not be well-tolerated, and thus either decrease the dosage to the previous level or test the same dose in a greater number of participants to verify the absence of further DLTs.

Designing dose escalation protocols in phase I clinical trials

A dose escalation protocol common to phase I trials should be based on a set of guidelines indicating how to proceed with the dose escalation according to how participants respond to the current dose level. Besides defining the number of patients per cohort/dosage level, the protocol should specify:¹

When to increase the dose of the novel drug to the next level (i.e., what DLT rate indicates to proceed to the next dosage?)
When to decrease the dose back to the prior level
What outcome (DLT rate, or otherwise) indicates/defines that the MTD has been reached

Most phase I clinical trials take a similar approach to determine tolerability, through smaller single-group/single-arm studies with healthy participants, which helps researchers be more certain that any side effects or adverse events are due to the novel drug. However, some trials, like those for cancer therapies, prefer to use patients with the condition under study.

Apart from the dose escalation protocol to be followed, other things sponsors need to keep in mind include the following:²

What are the appropriate (most relevant) endpoints to monitor?
How many dose levels should be assessed?
How should different DLT severities be graded?
What should the dose-limiting toxicity assessment ‘window’ be (i.e., for how long are patients monitored after dosing)?
What is the DLT target rate which will dictate the dosage that is defined as acceptable?
How many participants are needed?
What kinds of participants are needed (i.e., healthy volunteers vs. patients with a given condition)?
Under what circumstances/when should the trial be stopped?

Examples of dose escalation procedures

While there are a variety of procedures for dose escalation, the most prevalent is the 3 + 3 design. This approach is the most popular because it is relatively easy to implement, safe, adaptive, and provides statistically significant data. The main features of this rule-based procedure are given below:1

Three participants are enrolled and given the starting dose, which itself is extrapolated from animal studies
If there are no DLTs, the next three participants are enrolled and given the next dose level
If none experience a DLT, three more are enrolled and given the next higher dose
This cycle repeats until at least one participant experiences a dose-limiting toxicity (DLT)
If there is one DLT amongst the 3 participants, the escalation is paused and the same dose will be given to a new cohort of 3 participants
The dose escalation is stopped when 2 out of a cohort of either 3 or 6 patients experience a DLT (typically; again this depends on the specific protocol)
The MTD is then defined (in this case) as the dosage immediately below/before the level that resulted in 2 DLTs

Consider the following example, for a trial testing a novel drug, AABB. Data from prior preclinical studies on animals in a laboratory were extrapolated to suggest a starting dose in humans of 10 mg. The dosage levels are defined according to FIbonacci levels, according to common practice.1 The dose escalation protocol is defined as follows:

1. A cohort of 3 participants is enrolled and given the current dose (starting dose of 10 mg) of AABB.

If none (0/3) experience a DLT, enroll 3 new participants and administer the next dosage level (20 mg of AABB in this case)
If there are no DLTs, repeat from step (1) at the next dose level

If one participant (1/3) experiences a DLT, enroll 3 new participants and administer the same dosage, 20 mg

If none experience a DLT, proceed to the next dosage level, from step 1.

If 1 more experiences a DLT (2 out of 6 in total), stop dose escalation and define MTD as previous dosage

If 2+ more (3 or more out of 6 total) experience a DLT, drop dose to prior level and repeat from step 1

Challenges with defining and measuring DLTs

There are challenges and inconsistencies associated with defining and measuring DLTs, which can make it difficult to establish protocols for dose escalation studies and to ensure patient safety:¹,³

Lack of a standardized definition of DLT in clinical research. For cancer-related trials, there is a framework known as the Common Terminology Criteria for Adverse Events (CTCAE) classification which helps researchers classify DLTs of different severities. However, a general cross-discipline definition and standardization of DLTs in the clinical research industry does not exist - this has led to inconsistencies and can be a source of potential bias in determining what level of side effects are acceptable.
Novel drugs may not exhibit classical positive correlations between dose and efficacy or between dose and side effects. In other words, the practice of establishing the MTD just below the dosage producing a high number of DLTs may expose patients to unnecessary risk in the case of drugs that may be equally effective at even lower doses. This has been demonstrated, for example, in the case of immunostimulatory agents for anticancer therapy, which often have a wide therapeutic window.³ In that case, other indicators of efficacy could serve better to identify the dose for phase II trials, or could be combined with MTD data to provide a more complete picture of the dose-safety-efficacy relation.
Certain DLT definitions are limited to severe and life-threatening toxicities. The problem with this is that it doesn’t account for moderate toxicities, which can still severely affect participants’ quality of life and may become intolerable if they persist for long. The same is the case for side effects that are delayed or cumulative, which may not be observed within the DLT assessment window.
Dose escalation studies often administer novel drugs in fixed doses that are not adjusted to trial participants’ individual differences in aspects such as height, weight, ethnicity, etc. Each of these factors can (sometimes significantly) affect a drug's pharmacodynamics and tolerability. The relevance of this is still a matter of intense debate within the clinical research community, as accounting for such individual differences would greatly increase the complexity of phase I dose escalation studies.

Conclusion

Dose-limiting toxicity, or DLT, is used as an indicator of tolerability during dose-escalation studies, which are a common type of phase I clinical trial intending to identify the maximum tolerable dose (MTD) for subsequent trials. There are various possible dose-escalation protocols that establish rules for either increasing or decreasing the dose of the novel drug based on the number or rate of DLTs in each patient cohort.

While traditional dose escalation procedures tend to work relatively well for drugs that increase in efficacy as dose increases, this is not necessarily the case for all drugs and thus dose-escalation studies may not be a one-size-fits-all strategy. Furthermore, the lack of a standardized definition of what constitutes a DLT (and for classifying the varying severity of DLTs) makes it harder to maintain consistency across trials and also to safeguard patient safety. In cancer research, the CTCAE provides a framework for classifying adverse events and different levels of severity of DLTs for novel cancer drugs. This represents one example of collaboration in the field to enhance research efforts and enhance patient safety, which ultimately benefits patients and researchers alike.