Taking a closer look at the evidence
By Hans Herfarth, MD, PhD, AGAF
The debate of therapeutic drug monitoring (TDM) in the setting of anti–tumor necrosis factor (TNF) therapy has been ongoing for more than a decade. Reactive TDM, the measurement of drug concentrations in the context of loss of treatment response, is now generally accepted and recommended in multiple national and international inflammatory bowel disease (IBD) guidelines. Proactive TDM, defined as the systematic measurement of drug trough concentrations and anti-drug antibodies with dose adaptations to a predefined target drug concentration, seems to offer a possibility to stabilize drug levels and prevent anti-drug antibody formation due to low systemic drug levels, thus potentially preventing the well-known loss of response to anti-TNF therapy, which occurs in more than 50% of patients over time. However, proactive TDM is not endorsed by evidence-based guidelines, dividing IBD physicians into believers and nonbelievers and limiting uptake into clinical practice.
Dr. Hans Herfarth
As with reactive TDM, one should assume that the framework for proactive TDM should have been reliably established based on factual data derived from prospective controlled studies and not rely on retrospective cohorts or “Expert Panel” consensus statements. And indeed, several prospective controlled studies with sizable IBD patient cohorts have been published. Of note, all TDM studies for IBD were conducted in patients on anti-TNF maintenance therapy, and currently no prospective studies in larger IBD populations are available for proactive TDM during induction therapy. Two prospective studies, the PRECISION and the NOR-DRUM trial, report that proactive TDM is better than no TDM at all.
However, in the comparison of proactive TDM and reactive TDM (including at least one drug adaptation in maintenance or drug escalation based on clinical symptoms or biomarkers), three studies have demonstrated no significant differences in drug persistence or overall maintenance of clinical remission. Only a fourth (the pediatric PAILOT study) reported a lower frequency of mild flares and less steroid exposure in the proactive TDM arm over 1 year, but it did not show differences in drug persistence or overall clinical remission compared with the reactive TDM arm. Interestingly, the differences in flare frequency were apparent only in patients on monotherapy but not in the subgroup on combination therapy with an immunomodulator, stressing the well-known beneficial effect of a combination therapy with thiopurines in CD first shown in the SONIC trial.
One problem, at least in the TDM trials with infliximab (IFX), may have been a delay in optimizing IFX levels until the next drug infusion because of the turnaround time of the drug assays. However, even the most recently published ultra-proactive TDM study with ad hoc dose adjustments based on point of care testing of drug levels during maintenance IFX therapy did not result in a significantly lower failure rate of IFX therapy or differences in sustained remission compared with reactive TDM over 1 year (19% vs. 10%; P = .08 for IFX failure; 75% vs. 83%; P = .17 for sustained remission). Compared with reactive TDM, the proactive measurement of drug levels resulted in significantly more drug measurements (8.8 vs. 1 per patient per year) and consecutive drug adjustments. Of interest would be the assessment of cost-effectiveness based on these trial data.
The value of proactive TDM in induction therapy remains an ongoing concern. There is no doubt that the severity of intestinal inflammation with subsequent loss of drug in the intestine can result in low drug serum concentrations correlating to lower clinical responses and higher rates of immunogenicity with the formation of anti-drug antibodies. A recent study including patients with chronic immune-mediated inflammatory diseases such as IBD, rheumatoid arthritis, and psoriasis did not find a value in proactive TDM of IFX in the induction phase, but more severe IBD may have been underrepresented in this study. Administration of significantly higher induction dosing of adalimumab (160 mg weeks 0, 1, 2, and 3) with significantly higher trough levels compared with a standard induction regimen in the SERENE study has not been shown to increase the short- or long-term remission rates in UC or CD patients. Therefore, higher trough levels in a patient population do not automatically result in better outcomes, but proactive TDM may still have found a few patients who may have benefited from an even higher induction regimen. The UC and CD SERENE maintenance studies also evaluated proactive TDM versus clinical adjustment based on clinical and biomarkers. After 1 year, no differences in the efficacy endpoints of clinical, endoscopic, and deep remission were found. These somewhat surprising results, which have been reported only in meeting abstracts, suggest that simply increasing trough levels to a higher target (one of the primary aims of proactive TDM) is not an effective universal approach for achieving higher remission rates in induction or better outcomes in maintenance. Instead, the SERENE data show that similar results can be achieved by regular clinical follow-up and monitoring of loss of response based on symptoms and/or biochemical markers followed by drug adaptation (which may then also be based on reactive TDM).
