Prediction of response to remission induction therapy by gene expression profiling of peripheral blood in Japanese patients with microscopic polyangiitis

Microscopic polyangiitis (MPA), which is classified as an anti-neutrophil cytoplasmic antibody (ANCA)-associated small vessel vasculitis [1], is one of the most frequent primary vasculitides in East Asia. In Japan, Ozaki et al. conducted a prospective clinical trial on newly diagnosed patients with MPA in which they were administered the remission induction therapy according to disease severity [2]. Briefly, the MPA patients were stratified into three categories based on disease severity, namely mild form, severe form, and most severe form. The mild form included patients with slight disorder of one or more organs, a renal-limited type (except rapidly progressive glomerulonephritis (RPGN)), and a pulmonary-limited type (except pulmonary hemorrhage). The severe form consisted of patients with a generalized type (MPA with involvement of more than two organs), pulmo-renal type (glomerulonephritis plus either limited pulmonary hemorrhage or extended interstitial pneumonia), and RPGN. The most severe form included patients with diffuse alveolar hemorrhage, intestinal perforation, acute pancreatitis, cerebral hemorrhage, or concurrent presence of anti-glomerular basement membrane antibodies. After the establishment of diagnosis, the patients were treated according to the following protocols. 1) The mild form: low-dose corticosteroids (0.3–0.6 mg/kg/day) were administered; oral immunosuppressive agents (cyclophosphamide or azathioprine, 0.5–1.0 mg/kg/day or 25–75 mg/day, respectively) were optional. 2) The severe form: high-dose corticosteroids (0.6–1.0 mg/kg/day) and oral cyclophosphamide (0.5–2.0 mg/kg/day) were given; intravenous methylprednisolone (0.5–1.0 g/day for 3 days) was considered as an alternative. Instead of oral administration, the use of intravenous cyclophosphamide (0.5–0.75 g/m² monthly) was also allowed. 3) Most severe form: plasmapheresis (2.0–3.0 L/day for 3 days) was employed together with the regimen for the severe form described above.

After 18 months, the outcome of 47 patients, comprising of 22 with mild disease, 23 with severe disease, and 2 with the most severe disease, were analyzed. Remission, which is defined as the absence of clinical manifestations of active vasculitis (Birmingham Vasculitis Activity Score 2003: 0 or 1 point), was achieved in 42 out of 47 patients (remission rate, 89.4%). Among the 42 patients, 8 showed relapse of the disease, which is defined as the recurrence of at least one manifestation of vasculitis (recurrence rate, 19.0%). Ultimately, 5 out of 47 patients died (mortality rate, 10.6%).

These results demonstrate that the suggested therapeutic protocols are applicable for patients with MPA, but the possibility of relapse is indicated and, in the worst case scenario, death may occur regardless of the treatment. We have considered that, if response to the remission induction therapy would be predicted at an early point during the therapy, meticulous follow-up or application of additional regimens to the treatment could expectedly improve the outcome. For this purpose, we focused on the typical alteration of gene expressions after treatment in “good responders” who were conducted to persistent remission. We considered the prediction of “poor responders” who would not be conducted to remission or would relapse after remission when such typical alteration of gene expressions was not observed in the peripheral blood. Based on the results obtained through the transcriptome analysis, we nominated 16 genes from the comprehensive 8793 genes as predictors of response to remission induction therapy in MPA [3]. The 16 predictors included interferon (IFN) regulatory factor (IRF)7, IFN-induced protein with tetratricopeptide repeats (IFIT)1, IFIT5, 2′-5′-oligoadenylate synthetase-like (OASL), Charcot-Leyden crystal protein (CLC), guanylate binding protein 1 (GBP-1), proteasome (prosome macropain) subunit, beta type, 9 (PSMB9), hect domain and RLD (HERC)5, chemokine (C-C motif) receptor 1 (CCR1), CD36, membrane-spanning 4-domains, subfamily A, member 4 (MS4A4A), XIAP-associated factor-1 (BIRC4BP), phospholipid scramblase 1 (PLSCR1), defensin α1 and α3 (DEFA1/DEFA3), defensin α4 (DEFA4), and collagen type IX α2 (COL9A2).

In the present study, we determined the accuracy of prediction with the use of the aforementioned 16 predictors on 39 MPA patients who were selected randomly and retrospectively from the Japanese nationwide RemIT-JAV-RPGN cohort [4].

Lyons and colleagues reported that the transcriptome analysis of leukocyte subsets before treatment enabled the identification of gene signatures of ANCA-associated vasculitis, including MPA [8]. Similarly, McKinney et al. reported that the transcription signature of CD8⁺ T cells before treatment could predict prognosis in autoimmune diseases, including MPA [9]. On the contrary, we have focused on the altered gene expression in the peripheral blood of MPA patients between the period before and 1 week after the beginning of remission induction therapy. Therefore, it would seem likely that the prediction was based on the effects of the therapeutic reagents included in the remission induction regimen.

Among the 16 genes used as predictors in this study, the expressions of IRF7, IFIT1, IFIT5, OASL, CLC, GBP-1, PSMB9, HERC5, CCR1, CD36, MS4A4A, BIRC4BP, and PLSCR1 were decreased, whereas those of DEFA1/DEFA3, DEFA4, and COL9A2 were increased after treatment in “good responders” [3]. The relation between the decrease in several IFN-related genes, such as IRF7 [10], IFIT1 [11], IFIT5 [12], OASL [13], and GBP-1 [14], after the anti-inflammatory immunosuppressive treatment and the “good response” could be profound. ANCA-associated vasculitis, including MPA, has not been regarded as a type 1 IFN-driven disease [8]. We consider the expressions of the IFN-related genes as very sensitive to immunosuppressive treatment. Thus, we assume that the alteration can be a good marker of the therapeutic effects, though the expressions are not signatures of the disease.

In addition, we noted that the decrease in CLC gene expression after treatment reflected the “good response.” CLC proteins are mainly expressed in eosinophils [15], and the eosinophil count is rapidly reduced by corticosteroid treatment; therefore, the reduction of eosinophils in the peripheral blood after treatment possibly predicts “good response.” However, we could not assess the relation between prediction and alteration of leukocyte counts or rates of subpopulations in the peripheral blood after the treatment. This is the limitation of this study. We hope to challenge the clarification of this issue in our future studies.

We have demonstrated that the response to remission induction therapy can be predicted by monitoring the altered expressions of the 16 predictors in the peripheral blood at an early point of treatment in MPA patients. On the other hand, the accuracy tends to decline according to the increase in disease severity, though there is no statistical power. A significantly bigger sample size can elucidate the benefits and disadvantages of the prediction in future studies.

Although improvement to increase the sensitivity to predict “poor responders” is needed, such prediction at an early point of treatment, if any, should be significant. This prediction can help us consider meticulous follow-up or application of additional regimens to the treatment for the patients who are predicted as “poor responders.”

The response to remission induction therapy can be predicted by monitoring the altered expressions of the 16 predictors in the peripheral blood at an early point of treatment in MPA patients.