A prospective cohort study to determine the relationship between serum infliximab concentration and efficacy in patients with luminal Crohn's disease

BACKGROUND

Patients with Crohn's disease (CD) may experience disease relapse on maintenance infliximab. Anti-drug antibodies likely contribute to loss of response, and serum infliximab levels likely correlate with efficacy.

AIM

To prospectively evaluate the relationship between trough serum infliximab concentration and disease activity.

METHODS

Adult patients (N = 327) with a diagnosis of CD who had received at least five consecutive infliximab infusions and who planned to receive at least two additional infusions were enrolled. The Crohn's Disease Activity Index (CDAI), serum infliximab, C-reactive protein (CRP) and antibodies-to-infliximab (ATI) were assessed at baseline, week 4 and week 8. Receiver operating characteristic (ROC) analysis examined the relationship between infliximab concentrations and disease activity.

RESULTS

The mean CDAI score, which decreased 1.05 points between infusions, did not correlate with the mean change in trough infliximab concentration (+0.39 μg/mL; r = 0.099, P = 0.083), but was associated with the mean change in CRP concentration (r = 0.19, P < 0.001). Trough infliximab concentrations below 2.8-4.6 μg/mL best predicted a ≥ 70 point increase in the CDAI between infusions, and those below 2.7-2.8 μg/mL best predicted CRP >5 mg/mL at the second infusion. ATI at either visit decreased the proportion of patients with therapeutic infliximab trough levels compared with patients who were ATI negative (17.5% vs. 77.3% at visit 1 and 13.8% vs. 75.6% at visit 3; P < 0.001 for both comparisons).

CONCLUSIONS

This prospective study confirms the relationship between trough infliximab concentrations, inflammation and antibodies-to-infliximab. Infliximab trough concentrations below 3 μg/mL may increase the likelihood of symptoms and inflammation (ClinicalTrials.gov identifier: NCT00676988).

The relationship between stool hardness and stool composition in breast- and formula-fed infants

"Constipation" and "hard stools" are associated with formula feeding of both term and preterm infants and, in the latter, can lead to life-threatening complications. This study tested the hypothesis that stool hardness is related to excretion of fatty acid (FA) soaps in term infants, and in the extreme to milk bolus obstruction in premature infants. Stools (n = 44) were collected from 20 formula-fed and 10 breast-fed infants aged 6 weeks and were classified using visual charts for stool hardness on a 5-point scale (1, watery; 5, hard). Stools were analysed for nitrogen, minerals, and lipid, the latter divided between the soap and nonsoap fractions. We explored the relationship between stool hardness or solids content and stool constituents, relative to both wet and dry weight. Calcium and FA soaps were the dominant factors significantly related to stool solids and hardness score across the breast- and formula-fed groups. An 8% increase in stool dry weight FA soap content corresponded to a 1-point change in stool hardness score. Stools from formula-fed infants had a higher solids content and were classified as significantly harder than those from breast-fed infants (hardness scores, 4.0 +/- 0.5 versus 2.6 +/- 0.7, mean +/- SD) and on both a wet- and dry-weight basis contained severalfold higher levels of minerals and lipid and considerably less carbohydrate. Differences in lipids between formula- and breast-fed infants' stools were due almost entirely to FAs (mainly C16:0 and C18:0) excreted as soaps (27.7 +/- 7.5% compared to 3.1 +/- 4.1% of dry weight), suggesting the groups differed markedly in their handling of saturated FAs. An inspissated stool sample from a premature infant requiring surgical disempaction of an obstructed small intestine was found to be enriched in FA and calcium relative to the preterm formula. FA soaps, predominantly saturated, accounted for one third of the stool dry weight. These data support the hypothesis that calcium FA soaps are positively related to stool hardness; we speculate that this may, at least in part, explain the greater stool hardness in formula- versus breast-fed infants and milk bolus obstruction in preterm infants. This conclusion is consistent with the physical properties of calcium FA soaps.

Randomised trial of a ready-to-feed compared with powdered formula

Forty three infants were assigned randomly to a ready-to-feed infant formula or a standard formula that required reconstitution from powder. Despite similar nutrient composition of the two formulas those fed the powdered formula had significantly increased body weight and skinfold thickness gains, and became significantly heavier than a further group of 20 breast fed infants by 3 and 6 months. Of those fed the powdered formula 6/19 had become overweight (above the 90th or 97th centile) by 6 months, whereas 1/19 fed the ready-to-feed product was overweight at this age. While differences in fat absorption might have been contributory, our data suggest that errors in reconstitution of formula from powder might be the main cause for the growth differences observed. If it is appropriate to take the breast fed infant as a model, infants fed ready-to-feed formula in this study showed a more physiological pattern of growth than those fed a standard formula reconstituted from powder. These results require replication using other formulas as the findings have potentially important implications for infant feeding.