New data were presented at the American Academy of Neurology annual meeting on the effect of oral cladribine on lymphocyte subsets in relapsing multiple sclerosis, providing new insights for evaluating the safety and efficacy of the drug. Oral cladribine is a cell-depleting disease-modifying drug, which was approved for use in Canada in November 2017.
Cladribine is a prodrug (2-chloro-2′-deoxyadenosine) that mimics the nucleoside deoxyadenosine. It enters cells via the purine nucleoside transporter (Liliemark J. Clin Pharmacokinet 1997;32:120-131), where it accumulates intracellularly due to its resistance to adenosine deaminase. Cladribine is phosphorylated by deoxycytidine kinase (DCK) to an active triphosphate deoxynucleotide, which inhibits DNA synthesis and repair; and is metabolized by the phosphatase 5′-nucleotidase. Lymphocytes are preferentially depleted due to high levels of DCK, which activates the prodrug, and low levels of the phosphatase that inactivates the drug (Leist & Weissert. Clin Neuropharmacol 2011;34:28-35). Cell depletion is largely confined to T and B lymphocytes, with lesser effects on natural killer cells and monocytes (Salvat et al. AAN 2009; abstract P09.105). (See also Cladribine: intermittent immunosuppression in MS, NeuroSens, September 21, 2017.)
Oral cladribine is administered in weeks 1 and 5 in years 1 and 2, with a cumulative dose of 3.5 mg/kg. Lymphocyte kinetics were analysed using data from the phase III ORACLE-MS trial in early MS, and the pivotal CLARITY trial and its extension (Soelberg Sorensen et al. AAN 2018; abstract P5.351). After Year 1 dosing, there was a rapid reduction in CD19+ B cells, with the nadir (81-84% reduction) occurring at week 13 (i.e. seven weeks after the last dose). B cell reconstitution occurred from weeks 24-48. There was a lesser degree of T cell depletion (CD4+, -55%; CD8+, -48%), with nadirs at week 13 and a slower rate of recovery of CD4+ T cells. NK cells showed transient depletion (-44% at week 13), with recovery by week 24.
A separate analysis examined lymphocyte counts for up to 312 weeks after oral cladribine dosing in the CLARITY extension and PREMIERE safety registry (Soelberg Sorensen et al. AAN 2018; abstract P5.364). The absolute lymphocyte count (ALC) returned to normal values in 75% of patients by week 144. CD19+ counts returned to threshold values by 7.5 months post-dosing. CD4+ values returned to 40% of baseline at week 120. CD8+ cells showed less depletion, with values never dropping below the lower limit of normal throughout the treatment course.
Effect on memory B cells
Over the past decade, there has been a growing interest in the role of memory B cells in CNS inflammation and neurodegeneration in MS. Memory B cells (CD19+/CD27+) are the main producers of the pro-inflammatory cytokines TNF-alpha and lymphotoxin (Duddy et al. J Immunol 2007;178:6092-6099), which may contribute to CNS tissue damage. A further suggestion is that in some MS patients, memory B cells in the periphery may induce proliferation of interferon-gamma-secreting T cells that are specific to myelin antigens such as MBP and MOG (Harp et al. Eur J Immunol 2010;40:2942-2956).
Two UK studies presented at AAN 2018 examined the impact of DMDs on B cell subsets. In the first study, the number and relative proportion of memory B cells were determined during treatment with cladribine (at week 44) and alemtuzumab (after 1 and 2 years) (Ceronie et al. AAN 2018; abstract P5.365). Both treatments significantly depleted the absolute number and relative proportion of memory B cells in treated MS patients compared to untreated MS patients and healthy controls.
Similarly, cladribine produced significant depletion of memory B cells in a study of a subcutaneous formulation (Ceronie et al. AAN 2018; abstract S52 007). Interestingly, while CD19+ B cells soon recovered to baseline, the number of memory B cells remained below normal for up to 12 months post-dosing. High levels of DCK (which activates cladribine) were found both in B cells and in germinal centre B cells, which form memory B cells. The authors concluded that cladribine creates a molecular mechanism for selectively targeting memory B cells.
The most common adverse effect with oral cladribine is lymphopenia, consistent with the drug’s mode of action. An integrated safety analysis of data from the ORACLE-MS study of early MS, the CLARITY trial and extension and the PREMIERE safety registry reported that the overall incidence of infections was similar with oral cladribine 3.5 mg/kg and placebo (Cook et al. AAN 2018; abstract P3.407). The incidence of herpes zoster infections was higher with cladribine. The rate of infections was higher during periods of Grades 3-4 lymphopenia (57.23 per 100 patient-years vs. 24.50/100 PY outside of those periods); most cases were treatable infections of the upper respiratory tract. Four cases of herpes zoster occurred during periods of Grade 3-4 lymphopenia (incidence 4.50/100 PY vs. 0.73/100 PY outside of those periods); all cases were dermatomal and mild-to-moderate in severity.
To minimize severe lymphopenia, it is recommended that oral cladribine be initiated only in patients with a normal lymphocyte count. Patients should have no greater than Grade 1 lymphopenia prior to Year 2 dosing. The effectiveness of these recommendations was examined in the subgroup of patients in the CLARITY trial + extension receiving two courses of cladribine 3.5 mg/day over four years (cumulative dose 7.0 mg/kg) who met the dosing criteria (n=167) (Cook et al. AAN 2018; abstract P5.370). Grade 3 lymphopenia occurred in 1% at week 24 of year 1, and in 4% at week 24 in years 2-4. At the time of re-dosing (week 48 in the study), grade 3 lymphopenia was only observed in year 2 (incidence 1%). There were no cases of grade 4 lymphopenia at any time point.
A novel risk-benefit analysis used published safety and efficacy data from phase III studies to estimate the likelihood of help versus harm (LHH) with different disease-modifying therapies (Mitsikostas et al. AAN 2018; abstract P4.388). LHH was defined as the number needed to harm divided by the number needed to treat (NNH/NNT). NNH was based on the number needed to treat to have a dropout due to an adverse event, and NNT was based on the number needed to treat to achieve no evidence of disease activity (NEDA). The drugs compared were oral cladribine (CLARITY trial), dimethyl fumarate (DEFINE and CONFIRM), fingolimod (FREEDOMS and FREEDOMS II) and teriflunomide (TEMSO).
The NNT to achieve NEDA was 3.5 for oral cladribine, 7.3 for DMF, 4.7 for fingolimod and 11.6 for teriflunomide. The NNH was 68.5 for oral cladribine, 47.6 for DMF, 20.5 for fingolimod and 37.0 for teriflunomide. The estimated likelihood to help versus harm was 20.3 for oral cladribine, 6.5 for DMF, 4.3 for fingolimod and 3.1 for teriflunomide. This methodology for estimating the risk-benefit of treatments will require validation and would need to be supported by real-world data and clinical experience.