Predicting Effects of Adalimumab in Ulcerative Colitis
Predicting Effects of Adalimumab in Ulcerative Colitis
This single centre cohort study of adalimumab as a second line anti-TNF treatment in the treatment of ulcerative colitis shows that prior response to infliximab and higher early adalimumab serum concentrations at week 4 are predictive for long-term response.
This is the first large single centre study of ulcerative colitis patients treated with adalimumab after infliximab. In addition, our study is the first that includes pharmacokinetic data of previous infliximab treatment and adalimumab concentrations early after starting adalimumab.
The overall observed response rates in this cohort of ulcerative colitis patients treated with adalimumab as a second line anti-TNF agent were very similar to previously reported response rates from small open label series.
Our data show that prior infliximab response is an important factor in predicting response to the second line anti-TNF treatment. Although this can be intuitively considered logic, to our knowledge this was never demonstrated. However, we did still observe durable response to adalimumab in patients that were considered primary nonresponders to infliximab. Antibodies to infliximab and undetectable infliximab trough concentrations were detected in 13.5% of patients. These patients had all a durable responses to a second line anti-TNF. Unlike in another observational study, infliximab trough concentrations at discontinuation of infliximab did not correlate with response to adalimumab later. However in our study, many patients underwent infliximab dose optimisation prior to switching to adalimumab, potentially confounding the results. Other studies have confirmed durable responses when switching to a second line and even third line anti-TNF.
Like most other pharmacokinetic studies with anti-TNF monoclonal antibodies across indications, we confirm a clear relationship between serum drug concentration and clinical effect. In our study, adalimumab serum concentrations as early as week 4 after the standard high dose induction regimen (160 mg at week 0 and 80 mg at week 2) correlate with long-term efficacy. Also in accordance with other studies, we see a wide inter-individual variation in serum drug concentrations. Classic pharmacokinetic factors including body weight, BMI, but also albumin and CRP insufficiently explain this variation. We found only a weak association with albumin. This may be due to unlike infliximab adalimumab is not indicated for acute severe ulcerative colitis. Indeed the association between low trough levels and low albumin has only been described in for ulcerative colitis patients treated with infliximab and only for albumin levels below the normal range of values. We found a higher optimal cut-off for year one response compared to response at week 12. We speculate that early low levels are a risk factor for developing antibodies to adalimumab (ATA) causing loss of response over time. This in contrast to patients with early high(er) levels where immunogenicity (ATA formation) is probably suppressed.
As previously observed a high percentage of patients needed early dose escalation. Several authors have suggested that higher initial dosing, especially in ulcerative colitis, could improve early and long-term response. Studies to test higher doses of adalimumab are planned. Alternatively, more rational and potentially cost effective, an individualised dosing regimen based on disease burden rather than uniform dosing has recently been advocated. However, in ulcerative colitis we lack good biomarkers to guide individualised dosing. Faecal markers such as calprotectin could be a candidate marker in the future. On the other hand, high serum concentrations may be a consequence of mucosal healing as faecal loss may occur or higher concentrations mean that TNF-α has been adequately and sufficiently saturated.
Finally, no new safety signals except reiterating the need for thorough screening for dysplasia in this population with long-standing refractory ulcerative colitis.
This study has limitations inherent to the retrospective trial design. Although the serum samples have been collected prospectively, clinical data were collected retrospectively. In addition, patients have been treated in a noncontrolled fashion, including dose escalation based solely on clinical and endoscopic evaluation. This reflects the 'real life' situation and enabled studying different potential risk factors such as prior response to infliximab.
In conclusion, our study looking at adalimumab as a second line treatment in ulcerative colitis confirms similar results to first line open label series. We demonstrate that primary response to infliximab and higher early adalimumab concentrations predict short-term and long-term response. We speculate that early measurement of drug concentration and early intervention in patients with low serum concentrations may improve response rates and duration of response.
Discussion
This single centre cohort study of adalimumab as a second line anti-TNF treatment in the treatment of ulcerative colitis shows that prior response to infliximab and higher early adalimumab serum concentrations at week 4 are predictive for long-term response.
This is the first large single centre study of ulcerative colitis patients treated with adalimumab after infliximab. In addition, our study is the first that includes pharmacokinetic data of previous infliximab treatment and adalimumab concentrations early after starting adalimumab.
The overall observed response rates in this cohort of ulcerative colitis patients treated with adalimumab as a second line anti-TNF agent were very similar to previously reported response rates from small open label series.
Our data show that prior infliximab response is an important factor in predicting response to the second line anti-TNF treatment. Although this can be intuitively considered logic, to our knowledge this was never demonstrated. However, we did still observe durable response to adalimumab in patients that were considered primary nonresponders to infliximab. Antibodies to infliximab and undetectable infliximab trough concentrations were detected in 13.5% of patients. These patients had all a durable responses to a second line anti-TNF. Unlike in another observational study, infliximab trough concentrations at discontinuation of infliximab did not correlate with response to adalimumab later. However in our study, many patients underwent infliximab dose optimisation prior to switching to adalimumab, potentially confounding the results. Other studies have confirmed durable responses when switching to a second line and even third line anti-TNF.
Like most other pharmacokinetic studies with anti-TNF monoclonal antibodies across indications, we confirm a clear relationship between serum drug concentration and clinical effect. In our study, adalimumab serum concentrations as early as week 4 after the standard high dose induction regimen (160 mg at week 0 and 80 mg at week 2) correlate with long-term efficacy. Also in accordance with other studies, we see a wide inter-individual variation in serum drug concentrations. Classic pharmacokinetic factors including body weight, BMI, but also albumin and CRP insufficiently explain this variation. We found only a weak association with albumin. This may be due to unlike infliximab adalimumab is not indicated for acute severe ulcerative colitis. Indeed the association between low trough levels and low albumin has only been described in for ulcerative colitis patients treated with infliximab and only for albumin levels below the normal range of values. We found a higher optimal cut-off for year one response compared to response at week 12. We speculate that early low levels are a risk factor for developing antibodies to adalimumab (ATA) causing loss of response over time. This in contrast to patients with early high(er) levels where immunogenicity (ATA formation) is probably suppressed.
As previously observed a high percentage of patients needed early dose escalation. Several authors have suggested that higher initial dosing, especially in ulcerative colitis, could improve early and long-term response. Studies to test higher doses of adalimumab are planned. Alternatively, more rational and potentially cost effective, an individualised dosing regimen based on disease burden rather than uniform dosing has recently been advocated. However, in ulcerative colitis we lack good biomarkers to guide individualised dosing. Faecal markers such as calprotectin could be a candidate marker in the future. On the other hand, high serum concentrations may be a consequence of mucosal healing as faecal loss may occur or higher concentrations mean that TNF-α has been adequately and sufficiently saturated.
Finally, no new safety signals except reiterating the need for thorough screening for dysplasia in this population with long-standing refractory ulcerative colitis.
This study has limitations inherent to the retrospective trial design. Although the serum samples have been collected prospectively, clinical data were collected retrospectively. In addition, patients have been treated in a noncontrolled fashion, including dose escalation based solely on clinical and endoscopic evaluation. This reflects the 'real life' situation and enabled studying different potential risk factors such as prior response to infliximab.
In conclusion, our study looking at adalimumab as a second line treatment in ulcerative colitis confirms similar results to first line open label series. We demonstrate that primary response to infliximab and higher early adalimumab concentrations predict short-term and long-term response. We speculate that early measurement of drug concentration and early intervention in patients with low serum concentrations may improve response rates and duration of response.