Lenabasum for skin disease in patients with diffuse cutaneous systemic sclerosis

Spiera et al. report promising results of a small multicenter, double-blind, randomized, placebo-controlled, phase II study of lenabasum, a cannabinoid type 2 (CB2) receptor agonist, for the treatment of skin disease in patients with early diffuse cutaneous systemic sclerosis (dcSSc) on stable background immunosuppression. This study is important because it examines the effects of cannabinoid (CB) modulation in SSc, a hitherto unexplored pathway, it uses the American College of Rheumatology (ACR) Combined Response Index in Systemic Sclerosis (CRISS) index as an outcome in a 12-week as opposed to its intended use in 12-month clinical trials, and it redemonstrates the utility of obtaining skin biopsies to assess secondary outcomes in SSc clinical trials.

Two CB receptors have been identified, the type 1 (CB1), identified in 1990, and the type 2 (CB2), identified in 1993. At the earliest stages of embryogenesis during the three-layer blastocyst phase, CB receptors are present on a variety of cell types including immune cells as well as microglia, adipocytes and intestinal cells (Figure) [1–3]. As opposed to the CB1 receptors that are primarily located in the central nervous system where stimulation results in analgesia and behavioral changes, CB2 receptor stimulation primarily regulates immune cell function including T-lymphocyte subset balance, cytokine production and cell migration (Figure). More recently, CB2 receptors also have been discovered on brainstem neurons [4].

Figure.

Cannabinoid receptors in systemic sclerosis. Cannabinoid (CB) receptors are present in the earliest stages of embryogenesis where they coordinate cell proliferation and differentiation. In adults, type 1 (CB1) receptors are primarily expressed in the brain while type 2 (CB2) receptors are expressed in many tissue types that are derived from all three germ layers.

As the authors highlight, preclinical work revealed overexpression of CB2 receptors in cultured dermal fibroblasts from patients with SSc compared to healthy control participants and a reduction in transforming growth factor-beta (TGF-β) and collagen synthesis in response to lenabasum, providing a potential mechanism of action for lenabasum’s therapeutic effects. The background for this study exemplifies how translational research using tissues from patients with systemic sclerosis (SSc) can provide scientific rationale for clinical trials. To this end, two skin biopsies were obtained before, and at study conclusion. Clinical improvement in lenabasum-treated patients mirrored decreases in skin inflammatory infiltrates on histology and expression of genes related to inflammation and fibrosis. These results demonstrate the utility of skin biopsy collection to provide insights into mechanism and evaluation of the performance of new SSc outcomes. The next logical step may be to identify minimal biologically important differences in histological assessments and gene expression in SSc skin biopsies akin to the minimal clinically important differences in the modified Rodnan skin score (mRSS) or patient-reported outcome instruments. Lenabasum’s effects upon tissue-level phenotypes, such as hyalinized collagen score [5] or alpha smooth muscle actin staining [6] as well as the recently described computer vision-generated SSc dermal Fibrosis Score [7] may provide additional key insights into the therapeutic role of CB2 receptor stimulation in SSc dermal fibrosis. Moreover, two SSc skin gene expression biomarkers namely, the four-gene biomarker [8] and the Scleroderma Skin Severity Score (4S) [9] have been published, and prospective studies provide an opportunity to test the responsiveness of these molecular outcomes. Validation of quantitative molecular outcomes that are associated with improvement in survival or with the way a patient feels or functions would improve SSc clinical trial design.

Because of broad expression and pleiotropic effects of CB2 receptors throughout the body including the intestines and the vasculature (Figure), the effects of lenabasum upon gastrointestinal disease and Raynaud severity in SSc patients would have been interesting to learn. It is conceivable that CB2 activation would impact these organ systems, and future studies should consider including the Raynaud Condition Score as well as the Gastrointestinal Instrument (UCLA-SCTC GIT 2.0) or similar outcome instruments to assess symptom burden. Similarly, investigation of the role of lenabasum in patients with limited cutaneous SSc (lcSSc) should be considered. Although patients with the limited subset of SSc account for 60-65% of SSc patients, virtually no clinical trials include these patients leading to significant knowledge gaps and a dearth of effective treatment. Given the presence of CB2 receptors on adipocytes that may contribute to SSc dermal pathogenesis through mesenchymal to epithelial transition, lenabasum treatment should be considered for broader use in the SSc patient population where loss of intradermal and subcutaneous fat is a disease feature [10].

The decision to include CXCL4 and C-reactive protein (CRP) and interleukin-6 (IL-6) as serum biomarkers warrants discussion. The CRISS, the primary outcome for the lenabasum trial, includes mRSS and forced vital capacity (FVC %) predicted measurements, so it is rational to include blood biomarkers that have been previously shown to be associated with skin and lung disease in SSc. CXCL4 was shown in prior studies to be associated with skin and lung fibrosis as assessed by the mRSS and chest high-resolution computed tomography and FVC % predicted <70, respectively, in 779 patients with SSc (41% dcSSc) [11]. Ross et al., studied 1,545 patients with prevalent SSc (25% dcSSc) enrolled in the Australia Scleroderma Cohort and found that CRP≥5 mg/L was associated with baseline mRSS > 20 and an FVC < 80% predicted (both p<0.05) while a French national registry of 625 SSc patients (29% dcSSc) with incident SSc (<3y SSc disease duration) showed a CRP > 8 mg/L was associated with dcSSc and poorer survival over a median (Interquartile range [IQR]) follow-up of 4.4 (5.3) years [12, 13]. Levels of interleukin-6 (IL-6) produced by lymphocytes and macrophages have been shown to be associated with increased skin fibrosis, and with prevalent SSc-interstitial lung disease and its progression [14]. The authors report that mean serum levels of CXCL4, CRP and IL-6 were within normal limits at baseline and did not change during the trial. Whether any patients had elevated values and whether those patients also demonstrated higher expression of genes related to inflammation is also important information because it supports the notion of subtypes or endophenotypes within the broader dcSSc subset. Investigation of plasma Krebs von den Lungen (KL-6) might have also been interesting to pursue as elevated levels have been reported to be predictive of fibrotic lung involvement and progression in patients with SSc [15].

As mentioned above, although the primary trial outcome was the CRISS, the authors report mRSS changes in two ways. First, in the results section, they report the greatest treatment effect of lenabasum was −2.6 (standard error/SE=1.9), p=0.09 (for one-sided mixed effect model with repeated measures/MMRM test) and p=0.17 for a two-sided MRMM test) at the last time point evaluated. In their discussion, the authors mention a mean improvement from baseline mRSS in the treated group of −4.6. For clarity, the 2-point mRSS reduction in the placebo-treated patients subtracted from the 4.6-point mRSS reduction in the lenabasum treated patients is how the authors derive the −2.6-point mRSS lenabasum treatment effect. It is important to note that the study was not powered to assess change in mRSS with lenabasum treatment, and the results must be interpreted in that context.

Patients completed the Patient-Reported Outcomes Measurement Information System 29-item General Health Profile (PROMIS-29) at baseline and weeks 4 and 12 and results are provided for two (physical function and ability to participate in social roles and activities) of the possible seven health-related quality of life (HRQoL) domains. Prior work has demonstrated the responsiveness to change of PROMIS-29 in SSc patients over 12 months of follow-up with strong correlations between the PROMIS-29 physical functioning subscale and HAQ-DI and the SF-36 physical component score, and moderate correlations between PROMIS-29 anxiety, depression, and fatigue subscales and the SF-36 Mental Component Score [16]. Inclusion of PROMIS-29 in a larger and longer trial will permit evaluation of all seven domains. The effects of lenabasum upon fatigue (19% in lenabasum arm vs. 7% in placebo arm), as well as anxiety and depression will be important to determine in light of reports of CB2 receptors in the brain [4].

Although oral methotrexate and mycophenolate mofetil treatment have been shown to improve or be associated with improvement in SSc skin disease as assessed by the mRSS, both are associated with an increased risk of infection. Moreover, concern about methotrexate with its attendant risk for lung toxicity in patients with underlying interstitial lung disease limits its broad use. The study results show no signal for increased risk of infection in the lenabasum group although the numbers are small and future larger and longer duration studies will be required to test this hypothesis. Importantly, the trial design where patients were randomized to placebo BID for 12 weeks; or one of two treatment arms (lenabasum 5 PO QD + placebo or lenabasum 20 mg PO QD + placebo for four weeks followed by lenabasum 20 mg PO BID in both groups for eight weeks) permitted evaluation of a dose effect for adverse reactions that were not observed. The most commonly reported adverse effects including dizziness and headache could result from non-specific CB1 receptor stimulation or from brainstem CB2 receptor stimulation, thus it would be interesting to know the specificity of lenabasum for CB2 versus CB1 receptors. Possibly the one patient that withdrew from the lenabasum treatment group due to severe diziness overexpresses CB2 CNS receptors. However, the similar occurrence of CNS side effects in both the active- and placebo-treated participants is reassuring.

The time from study start, October 12, 2015, to report, January 13, 2020, is notably long given involvement of investigators at nine academic centers with active scleroderma programs. It would be advantageous for patients, investigators and pharmaceutical companies alike if the scleroderma community could harmonize clinical trial recruitment to permit one or a few studies to be actively recruiting simultaneously. This synchronization will require agreement by all involved parties that may be difficult to achieve, but it would speed trial completion, dissemination of results and reduce clinical trial costs. Efforts are underway through the Scleroderma Clinical Trials Consortium to prioritize recruitment of dcSSc patients into clinical trials sequentially. The positive results herein will likely increase enthusiasm for hastened enrollment in the upcoming larger phase 3 trial.

Given that generic methotrexate and mycophenolic acid formulations are available, cost will be a major consideration in the decision to prescribe lenabasum for SSc skin disease. The high cost of nintedanib, the first FDA-approved drug for SSc, bars its use as first-line treatment in patients with progressive SSc-ILD. Benefits of lenabasum may include the lack of blood monitoring requirements and no increased apparent infection risk if these results are confirmed in subsequent studies. Given the improvement in mRSS and PRO instruments in the lenabasum treated group in the present trial, it may be ethical to conduct future randomized trials that compare lenabasum to methotrexate and mycophenolic acid formulations head-to-head. If lenabasum is shown to significantly improve skin fibrosis and possibly gastrointestinal motility and Raynaud phenomenon, then lenabasum could emerge as first-line treatment for SSc patients.