Testing Whether Precision Dosing Improves Clinical Outcomes

TDM feasibility trials can demonstrate that TDM-guided dose individualization is effective in optimizing drug exposure, but they are not designed to prove that TDM actually improves treatment outcomes. Nevertheless, these studies are essential for test-driving dose-adjustment protocols, workflows, and infrastructure, as well as for identifying unforeseen barriers to execution of dose-adjustment recommendations. Consequently, the parameters derived from TDM feasibility trials are critical for informing the design and implementation of subsequent randomized controlled trials (RCTs) for testing whether dose individualization improves treatment outcomes (Figure 1, Stage 4). Based on TDM feasibility trial data, much consideration during the design of the RCT should be given to defining the comparison groups and the criteria for inclusion and exclusion of subjects with respect to the population of patients most likely to benefit from TDM. For example, it may be prudent to select only those individuals who are over- or under-exposed as your study population,i.e. those at high risk of treatment failure due to lack of efficacy or toxicity, rather than comparing outcomes for “TDM” vs “no TDM” in all patients receiving the drug.
One possible concern related to conducting a randomized controlled trial comparing TDM with the standard of care (i.e. no TDM) is that some may consider it unethical to abstain from increasing the drug dose in control patients with low systemic exposure. The underlying assumption in such an argument is that increasing the drug dose in an under-exposed patient will improve the clinical outcome. Based on this assumption, however, it may further be argued that it would be unethical not to measure drug levels in all patients to identify the underexposed individuals who would benefit from dose increases, and the ethical solution may be to apply TDM universally.33 A “catch-22” of sorts, this assumption is exactly what the randomized controlled trial in attempting to prove! Without RCTs demonstrating clear benefits of TDM versus no TDM for at least some oral small molecule anticancer drugs, the current oncology practice will not have enough incentive to change and all underexposed patients receiving the standard of care will continue to be prescribed sub-therapeutic doses of these drugs. For example, at our institution there are approximately 550 patients with active imatinib prescriptions. Despite evidence to suggest that approximately 150 to 350 of these individuals may be under-dosed (~25%-65%),29,36,37 our laboratory receives fewer than 3 samples per week for imatinib level monitoring.
To the best of our knowledge, there is no randomized controlled trial, either published in a peer-reviewed journal or ongoing, that uses a clinical outcome as the primary endpoint and demonstrates the benefit of TDM of targeted oral anticancer drugs. Nevertheless, results from several trials have been published in abstract form only, and thus may need to be considered with caution. Rousselot et al. reported the final results of the randomized OPTIM Imatinib study, which demonstrated that TDM increases the rates of molecular response in patients with chronic myeloid leukemia (CML).36 Another abstract from the same group reported that TDM of dasatinib resulted in reduced risk of pleural effusions and high molecular response rates in CML.38 Results of a trial by Gotta et al. were published in a peer-reviewed journal, but this study did not compare TDM versus no TDM. Instead, “routine TDM” was compared to “rescue TDM” for Imatinib in 57 CML patients.24 The primary endpoint was a combined outcome (failure- and toxicity-free survival with continuation on imatinib) over 1-year follow-up. This dose individualization trial could not demonstrate additional benefit of “routine TDM” using intention-to-treat analysis, due to a small number of study participants and, surprisingly, limited prescribers’ adherence to dosage recommendations. However, using as-treated subgroup analysis of “routine TDM” group, 10 of 14 of patients receiving the recommended dosage after one cycle of TDM remained event-free compared to 3 of 13 patients for whom dosage recommendations were not correctly adopted (P=0.033).
Thus, despite numerous studies (primarily in Stages 1-3 of Figure 1) providing evidence to suggest that TDM of oral small molecule anticancer drugs may be beneficial to patients, and perhaps even cost-effective,39 there seems to be no published study that definitively demonstrates such benefit. Historically, there are examples of drugs, such as calcineurin inhibitors, for which TDM became the standard of care without demonstrated superiority of “TDM” vs “no TDM” in RCTs. However, in the present culture emphasizing evidence-based medicine, having more drugs reach stage 4 of Figure 1 must be the goal, as this will encourage requests for TDM by medical oncologists as well as increase the chances for reimbursement of costs.