Increased Serum LIGHT Levels Correlate with Disease Progression and Severity of Interstitial Pneumonia in Patients with Dermatomyositis: A Case Control Study

Background Activated CD8+ T cells play an important role in the pathogenesis of dermatomyositis (DM) with interstitial pneumonia (IP). Serum CD8+ T-cell activator, LIGHT, and Th1/Th2/Th17 cytokines were measured in DM-IP patients and compared with clinical parameters to investigate their usefulness. Methods The correlations between the clinical findings and serum LIGHT and Th1/Th2/Th17 cytokine levels were investigated in 21 patients with DM-IP (14 with rapidly progressive IP [RPIP] and 7 with chronic IP [CIP], including 4 fatal cases of IP). Results The median serum LIGHT level was 119 (16–335.4) pg/ml, which was higher than that in healthy control subjects and DM patients without IP. The median serum IL–6 level was 14.7 (2.4–154.5) pg/ml (n = 13). The other cytokines were detected in only a few patients. The median serum LIGHT level in DM-RPIP patients (156 [49.6–335.4] pg/ml) was significantly higher than that in DM-CIP patients (94.3 [16–164.2] pg/ml) (P = 0.02). The serum IL–6 level did not correlate with either progression or outcome of DM-IP. ROC curve analysis determined a serum LIGHT level of ≥120 pg/ml to be the cut-off value for the rapid progression of DM-IP. Serum LIGHT levels correlated significantly with %DLco (R = 0.55, P = 0.04) and total ground-glass opacity scores (R = 0.72, P = 0.0002). The serum LIGHT level significantly decreased to 100.5 (12.4–259.3) pg/ml 4 weeks after treatment initiation (P = 0.04). Conclusions The serum LIGHT level may be a promising marker of disease progression and severity in patients with DM-IP.


Results
The median serum LIGHT level was 119 (16-335.4) pg/ml, which was higher than that in healthy control subjects and DM patients without IP. The median serum IL-6 level was 14.7 (2.4-154.5) pg/ml (n = 13). The other cytokines were detected in only a few patients. The median serum LIGHT level in DM-RPIP patients (156 [49. 6-335.4] pg/ml) was significantly higher than that in DM-CIP patients (94.3 [16-164.2] pg/ml) (P = 0.02). The serum IL-6 level did not correlate with either progression or outcome of DM-IP. ROC curve analysis determined a serum LIGHT level of 120 pg/ml to be the cut-off value for the rapid progression of DM-IP. Serum LIGHT levels correlated significantly with %DLco (R = 0.55, P = 0.04) and total ground-glass opacity scores (R = 0.72, P = 0.0002). The serum LIGHT level significantly decreased to 100.5 (12.4-259.3) pg/ml 4 weeks after treatment initiation (P = 0.04).

Ethics Statement
This study was approved by the ethical committee of Osaka Medical College (No. 1316) and complied with the guidelines of the Declaration of Helsinki. Written informed consent was obtained from each patient.

Treatment
The administration of 0.75-1.0 mg/kg/day prednisolone was begun in all patients and was concomitantly administered with cyclosporine (CSA) or tacrolimus (TAC). The physicians decided which of CSA or TAC was used. CSA was initiated at 4 mg/kg/day once a day before breakfast, and the CSA level was adjusted at 2 hours after administration (C2), to 1500 ng/ml or above [8,25]. TAC was initiated at 0.1 mg/kg/day twice a day before breakfast and dinner, and the trough level was adjusted to 10-20 ng/ml. The addition of IVCY was decided by physicians on the basis of the disease activity and patient condition.

Arterial blood gas analysis and pulmonary function test
Arterial blood gas analysis including PaO 2 , PaCO 2 , and the alveolar-arterial oxygen difference (AaDO 2 ) was performed on admission. Static and dynamic lung volumes were measured by spirometry (SYSTEM21; Minato Medical Science, Osaka, Japan). Vital capacity (VC) was determined by the N2 washout method. The diffusion capacity of the lung for carbon monoxide (DLco) was determined by the single-breath method. The results of pulmonary function testing are expressed as percentages of the predicted values.

HRCT scoring
HRCT was performed using an Aquilon 64-detector row CT scanner (Toshiba Medical System Corporation, Tokyo, Japan) on admission. Slice thickness was 1.0-1.5 mm at 10-mm increments, which included the entire lung. In DM-IP patients, the disease activity of IP correlates with the range of ground-glass opacity (GGO) [27]. Therefore, the area of GGO was scored to assess HRCT findings, as previously described [28]. The area of GGO in each lobe (right upper, middle, and lower, and left upper and lower lobes) of the lung was scored on a scale of 0-5 at the 3 sites: mid aortic arch, tracheal carina, and 1 cm above the top of the right diaphragm. The scores were summed as the total CT score.

Statistical analysis
Statistical analysis was performed using the Mann-Whitney U-test to compare median values and Fisher's exact test to compare frequencies. Correlations were evaluated by using Spearman's correlation coefficients. A value of P <0.05 was considered to indicate statistical significance. The data were analyzed using JMP software, version 11.0 (SAS Institute, Cary, NC, USA).

Levels of serum LIGHT and Th1/Th2/Th17 cytokines in DM-IP patients
As shown in Fig 1a and Table 2, the median levels (range) of serum LIGHT were 119 (16-335.4) in DM-IP patients (n = 19), 80.4 (39.9-90.2) in DM patients (n = 5), and 41.7 (21-97.2) pg/ml in the age-matched HC (n = 5). The serum LIGHT level in the DM-IP patients was higher than that in the DM patients (P = 0.06) and HC (P = 0.04). No significant difference was noted in the serum LIGHT level between the DM patients and HC. The median level (range) of serum LIGHT was 218 (49.6-335.4) pg/ml in the DM-IP patients with high CK (> 200 IU/l, n = 9) and tended to be higher than that of 105.6 (16-261.3) pg/ml in the DM-IP patients with normal CK ( 200 IU/l, n = 12) (P = 0.06). The median serum IL-6 level was 14.7 (2.4-154.5) pg/ml, and it was undetectable in 8 patients. The other cytokines were detectable in only a few patients.
Comparison of serological biomarkers of DM-IP between the DM-RPIP and DM-CIP groups ( Table 3). The respective median levels (range) of serum CRP and LIGHT were significantly higher in the DM-RPIP patients, 1.95 (0.29-12.8) mg/dl and 156 (49.6-335.4) pg/ml, than in the DM-CIP patients, 0.14 (0.04-0.63) mg/dl and 94.3 (16-164.2) pg/ml, (P = 0.0004 and 0.02) (Fig 1b). To determine the cut-off point effective for determining the progression of DM-IP, receiver operating characteristic (ROC) curve analysis was carried out on the serum LIGHT levels. The value that maximized the area under the ROC curve was 116.7 pg/ml for serum LIGHT (sensitivity: 78.6%, specificity: 64.3%). From this result, a LIGHT level of 120 pg/ml was determined as the cut-off value for rapid progression of DM-IP.

Comparison of serological biomarkers of DM-IP between the IP death and survivor groups
The anti-MDA5-Ab-positive rate was significantly higher in the IP death group (100%) than in the survivor group (25%) (P = 0.01), as shown in Table 4. No significant difference was noted    in the anti-ARS-Ab-positive rate or the CK, LD, CRP, KL-6, ferritin, LIGHT, and IL-6 levels between these two groups.

Correlation between serum LIGHT levels and other disease activity indicators of DM-IP
As shown in Table 5   Laboratory markers are presented as the median (interquartile range). DM, dermatomyositis; IP, interstitial pneumonia; Dead, dead due to IP; MDA5, anti-melanoma differentiation-associated gene 5; Ab, antibody; ARS, aminoacyl-tRNA synthetase; CK, creatine kinase; LD, lactate dehydrogenase; CRP, C-reactive protein; IL-6, interleukin 6. The P-values were estimated using Fisher's exact test or Mann-Whitney U-test. *P <0.05.

Discussion
The present study clarified that the serum LIGHT level was higher in the DM-IP than DM patients without IP and HC. In addition, the serum LIGHT level was significantly higher in the  DM-RPIP than DM-CIP patients. Moreover, the serum LIGHT level correlated significantly with %DLco and the GGO score on chest CT before treatment, and decreased after treatment. This is the first report to suggest the use of serum LIGHT as a possible marker of disease progression and severity in patients with DM-IP. Kurasawa et al. reported that activated Th1-type pulmonary T cells and CD25+CD8+ T cells play an important role in the development of corticosteroid-resistant IP in DM and polymyositis [29]. In DM-IP, lymphocyte infiltration in lung tissue, mainly that of CD8+ T and Th1 cells, is associated with the occurrence of an autoimmune response [30,31]. Comparison between chest HRCT and histopathological findings of the lung revealed a close correlation of GGO with inflammatory cell infiltration in the lung interstitium [32]. Our study showed that the level of LIGHT, which activates CD8+ T cells, and the GGO score on chest HRCT were closely correlated, suggesting that the serum LIGHT level reflects the expansion of activated CD8+ T-cell infiltration in the lung interstitium.
LIGHT induces marked activation of CD8+ T cells and promotes Th1 cytokine production [15]. Gono et al. reported that the IL-10 and TNF-α levels before treatment were high (6.6 [1.6-13.8] and 16.8 [10.6-29.3] pg/ml, respectively) in 38 patients with polymyositis/DM-IP [33]. In our study, the levels of many Th1/Th2/Th17 cytokines excluding IL-6 were lower than the limits of detection. This inconsistency may have been due to differences in patient background and the measurement method of the cytokines. A high KL-6 level, anti-MDA5 Ab positivity, high serum levels of ferritin, IL-6, IL-10, and IL-8, and a high level of AaDO 2 have been reported as serological factors in DM-IP that indicate a poor prognosis [33][34][35][36][37][38][39]. Sato et al. reported that anti-MDA5 Ab is a useful prognostic indicator of DM-IP [37]. In fact, the patients in all of the fatal cases in our study were anti-MDA5 Ab-positive. Nara et al. reported that the serum IL-6 level was 9 pg/ml or higher in all 6 patients with CADM-IP who died of IP, whereas it was lower than 9 ng/ml in 5 of 6 survivors [39]. It has also been reported that in patients with DM-IP, those with a serum ferritin level 1500 ng/ml, or those with a serum ferritin level 600 ng/ml plus AaDO 2 45 mmHg, have poor prognoses [8,33]. In this study, the AaDO 2 and serum ferritin, IL-6, and LIGHT levels were not associated with the outcome of DM-IP. This may have been due to the aggressive intervention with more potent immunosuppressive drugs from the early disease stage in patients with high levels of these markers.
Based on the findings obtained in this study, our recommended therapeutic strategy for DM-IP is shown in Fig 3. Among patients with DM-RPIP, for those with any of the poor prognostic factors of anti-MDA5 Ab-positivity, serum ferritin 600 ng/ml plus AaDO 2 45 mmHg, or serum LIGHT 120 pg/ml, combined administration of prednisolone at 1 mg/kg/ day and calcineurin inhibitor (CSA C2 level or TAC trough level adjusted to 1500 ng/ml or 10-20 ng/ml, respectively), and aggressive additional doses of IVCY are recommended as soon as possible after diagnosis.

Conclusion
We propose that the serum LIGHT level may be a promising biomarker of disease progression and severity in patients with DM-IP. This was a retrospective study involving a small number of patients. The investigation of additional accumulated cases may clarify the usefulness of LIGHT as a prognostic factor of DM-IP, and thus further study is desired.