Adverse events during the titration phase of interferon-beta in remitting-relapsing multiple sclerosis are not predicted by body mass index nor by pharmacodynamic biomarkers

Background This study aimed to correlate body mass index or biomarkers with the frequency of common adverse events (AEs) with subcutaneous IFN β-1a during treatment titration in patients with relapsing-remitting multiple sclerosis previously naïve to IFN β. Methods Eighty-four patients (66.3% females) were followed up during 8 weeks, 25.3% were overweight and 14.5% were obese. Results Biomarkers steadily increased during all study period by 45.3% for β2-microglobulin, 262.8% for olygoadenylate synthetase-1, and 92.8% for neopterin. Overall AE reporting did not vary with the dose or treatment duration. Conclusions BMI was not predictive of increased risk for AEs. Biomarkers did not discriminate on the frequency of any AE either.


Background
More than one decade ago, recombinant human interferon β (IFN β) was the first approved disease modifying drug, and still remains as a standard first-line therapy for relapsing-remitting multiple sclerosis (RRMS).
β2-microglobulin (β2MG) is associated with class I antigens of the major histocompatibility complex on the surface of lymphocytes and is considered to be a marker for disease activity in some immune and neoplastic disorders. Niezgoda et al. found a significant decrease in cerebrospinal fluid and serum β2MG levels in MS patients after cladribine treatment, associated with a slight but significant clinical improvement measured by EDSS [9]. However, Bagnato et al. followed the evolution of serum β2MG in untreated patients and failed to show any relationship between this marker and disease activity [4].
Neopterin (Np) is synthesized by macrophages upon stimulation with interferon and it is indicative of a proinflammatory immune status. Np serves as a marker of cellular immune system activation under the control of T helper cells type 1. Elevated Np concentrations are among the best predictors of adverse outcome in patients with HIV infection, in cardiovascular disease and in various types of cancer [10]. Most studies however have not found a clear correlation with multiple sclerosis progression [3,11].
Patients with MS tend to be more overweight than the general population [12] .Obesity seems to contribute to a delay in the diagnosis of MS through a still undetermined mechanism [13].
IFN-beta-1-a-induced immunomodulation in vivo depends on the administration schedule being 2-3 times greater the effect when a same weekly dose is divided in three injections [14].
Indicators of treatment response or drug-related AEs, either non-modifiable pharmacogenetic indicators like polymorphisms of genes or pharmacodynamic markers [15] might contribute to identify patients who can better benefit from treatment.
Recently, a study showed differences between responders and non responders in the genes associated with ion channels and signal transduction pathways. This study also suggests that genetic variants in heparan sulfate proteoglycan genes may be of clinical interest in MS as predictors of the response to therapy [16] .So far, no study has addressed a correlation of either body mass index (BMI) or IFN β biomarkers with the frequency of common AEs, which could help develop a more individualized IFN β therapy. The aim of the present clinical trial was to address whether variations of BMI or pharmacodynamic biomarkers are associated to different AE profiles during the titration phase of IFN β-naïve patients with RRMS.

Methods
This interventional open-label, single arm trial (EudraCT Number 2006-000606-23) was performed in 12 hospitals in Spain from July 2008 to June 2010, and was approved by all the Institutional Review Boards prior to initiation. All patients signed an informed consent form before enrollment, and were included in the study if aged between 18 and 60 years, diagnosed with RRMS, had suffered from one or more relapses within the previous 12 months, had an Expanded Disability Status Scale (EDSS) score < 5.5, and had not been previously treated with any IFN β or any other disease modifying drug (DMD). Exclusion criteria included progressive forms of MS, current or past liver abnormal function, leukocytopenia during the month prior to the start of the study, and presence of any other systemic disease.
The study treatment was IFN β-1a (Rebif®, Merck Serono), administered subcutaneously three times per week, starting with a dose of 8.8 μg for 2 weeks, then increasing up to 22 μg for other 2 weeks, and finally from day 28 on, at the targeted dose of 44 μg. The study end was after 56 days, following which patients were treated based on normal physician's practice. On days 1, 14, 28 and 56 blood samples were taken before the next administration of the study drug to perform a complete blood cell count, analysis of liver and renal function, and also determination of the three MS biomarkers: β2M was measured by chemiluminescence (Immulite2000, Siemens), Np by ELISA (DRG International) and OAS1 by RIA (Izasa). Adverse event logs were documented at each visit.
At the same time serum samples were stored frozen at −70C until shipped to a central laboratory for the determination of the three biomarkers β2 microglobulin, neopterin and 2'-5' oligoadenylate synthetase. β2 microglobulin was measured by chemiluminescence with reagents provided by Siemens. Neopterin was assessed by enzyme-linked immunosorbent assay (ELISA) using DRG International reagents. 2'-5' oligoadenylate synthetase was determined by means of radioimmunoassay with reagents from Izasa. An adverse event log was completed at each visit. The Student t-test for related and unrelated samples was used to compare means as appropriate. The correlation between BMI, biomarkers and lab results was also assessed using the Pearson or Spearman coefficients (for variables of normal or non parametric distribution according to the Kolmogorov-Smirnov test). The significance level was set at 0.05.
The means of the biochemical, hematological and biomarkers values will be compared to the baseline with a Student-t test for related samples so as to assess the evolution of these variables. The change in the biochemical, hematological and biomarkers values was also compared to the BMI by means of a scattergram and the Pearson (for variables of normal distribution) or Spearman (for non parametric variables) coefficients. The BMI mean of patients with and without headache, influenza-like syndrome and local toxicity with was compared with Student-t test for unrelated samples.
Likewise, the relationship between both the basal value of the three biomarkers and their change over time, among patient with or without adverse events was assessed with the Student-t test for unrelated samples. As for the relation between biomarkers and biochemical and hematological variables, their correlation by means of the Pearson or Spearman coefficients was also explored.
The primary study endpoint aimed to correlate the level of the three biomarkers, adjusted by BMI, and their association with the frequency of AEs. As a secondary endpoint, we aimed to further study these three biomarkers, adjusted by BMI, with the seriousness of AEs.
Baseline and final 56-day mean values of the different continuous variables studied are shown in Table 1.
Biomarkers also followed a significant (p < 0.01) and steady increase at 56 days after treatment compared with baseline values: β2M increased by 45.3%, OAS1 by 262.8% and Np by 92.8%.
As most common AEs, influenza-like symptoms were reported in 28 subjects (33.7%), headache in 15 (18.1%) and local reactions at the injection site in 11 subjects (13.3%). With regard to blood analysis, there were 8 cases of abnormal liver enzymes (9.6%), 4 cases of leukocytopenia (4.8%) and one case of thrombocytopenia (1.2%). Three (3.6%) patients were withdrawn from the study, 2 of them at second visit on day 14, due to severe chills and suicide thinking, respectively, and the third patient at third visit on day 28 due to a severe increase of transaminases.
Frequencies of AEs did not increase as the study progressed, despite dose escalation. The overall proportion of subjects reporting any AE was similar at each study visit (39.5% on day 14, 42% on day 28, and 42% on day 56).
No association between mean BMI and frequency of total or any specific AE was established. Similarly, no strong associations were found between BMI and variations of studied biomarkers (all correlations were under ±0,03). Additionally, no significant differences between either the mean baseline value or the variations of biomarkers were found to be associated to patients reporting any AE vs. those non reporting AEs (see Table 1).

Conclusion
In conclusion, we found that BMI and pharmacodynamic biomarkers were not predictive of an increased frequency of AEs. To our knowledge, these are the first results establishing a lack of correlation between BMI or biomarkers with frequency of AEs for IFN β-1a during the first two months of treatment. Therefore, an individualized approach should not be based on BMI or biomarkers from a safety perspective.