116
including internalization and phagocytosis with lysosomal degradation or endocy-
tosis and catabolism by the reticuloendothelial system [ 29 ]. The rate of degradation
is thought to be hastened by binding of anti-drug antibodies.
The importance of maintaining adequate drug levels was assessed in several
clinical trials for anti-TNF therapies, though interpretation of these results is often
challenging [ 30 ]. One primary concern is the early use of clinically oriented, and
also potentially subjective, outcome measures such as the CDAI. Another potential
pitfall is the type of assay employed to measure drug levels; for instance, conven-
tional ELISAs attempting to measure anti-drug antibodies are unable to do so in the
presence of drug. Interpretation of studies that incorporated nonscheduled dosing or
alternative dosing regimens from those that are the standard of care now, or did not
carefully quantify the impact of dose-modification that may have been allowed in
the study, also makes drug level interpretation challenging. Non-standardized tim-
ing of drug level measurement has been problematic as well.
Taking these caveats into consideration, several earlier clinical trials did attempt
to correlate the impact of infliximab drug levels with clinical response. In early trials
of episodic dosing, infliximab serum concentrations >12 μg/mL at 4 weeks after an
infusion were associated with a longer duration of response when compared to levels
<12 μg/mL (median 81.5 days vs. 68.5 days, respectively) [ 21 ], while another study
of episodic dosing appreciated higher rates of endoscopic healing and lower inflam-
matory markers such as CRP with increasing infliximab levels [ 23 ]. Similar results
have been appreciated in studies involving scheduled dosing as well. In a study
examining discontinuation of thiopurine therapy in those also receiving scheduled
infliximab and in clinical remission for 6 months or longer, higher infliximab levels
were correlated with lower CDAI, lower CRP, and, when comparing levels ≤2.23 μg/mL
to >2.23 μg/mL, a higher odds of requiring dose escalation (odds ratio (OR), 3.99
[95% CI, 1.53–10.11) [ 31 ]. In patients with UC, subgroup analyses of ACT 1 and 2
demonstrated that higher infliximab levels were associated with increased rates of
mucosal healing and clinical response to infliximab [ 32 ]. Seow and colleagues noted
that in a retrospective cohort of 115 patients receiving infliximab, rates of endoscopic
improvement and clinical symptom reduction were greater in individuals with detect-
able levels of infliximab compared to those with undetectable levels [ 24 ]. Undetectable
infliximab levels were also associated with a ninefold increased probability of subse-
quent colectomy in this study. Several large clinical trials in CD support the role of
infliximab level monitoring as well. In a post hoc analysis of data from the ACCENT
1 trial, median drug concentration was associated with clinical response as follows:
in patients with clinical response, median drug concentration was 12.9 compared to
8.8 during induction at week 6 and 4.6 compared to 1.9 at week 14 [ 33 ]. In fact, a
serum infliximab concentration >3.5 μg/mL at week 14 was associated with a 3.5-
fold increased odds of clinical response (95% CI 1.1–11.4). In the SONIC trials,
serum infliximab levels >3 μg/mL at week 30 were associated with increased rates of
mucosal healing at week 26 (OR 3.34, 95% CI 1.53–7.28) as well as corticosteroid-
free clinical remission at week 50 (OR 3.20, 95% CI 1.38–7.42) [ 34 ].
Recent observational data support the association between drug levels of inflix-
imab or adalimumab and rates of mucosal healing. In a cross-sectional study of 145
F.I. Scott and M.T. Osterman