Recombinant Human Soluble Thrombomodulin Treatment for Acute Exacerbation of Idiopathic Pulmonary Fibrosis: A Retrospective Study

Idiopathic pulmonary fibrosis (IPF) is an ultimately fatal fibrotic lung disease without identifiable etiology and characterized by a histologic pattern of usual interstitial pneumonia (UIP). Although the clinical course is usually chronic and slowly progressive, some patients deteriorate rapidly during the course of their illness; this is defined as an acute exacerbation of IPF (AE-IPF) [1]. AE-IPF was first described by Kondoh et al. [2] and is now well-recognized by physicians as a clinical event with high morbidity. Because of the paucity of studies, there are no definitive treatment guidelines for AE-IPF; however, corticosteroid (CS) pulse therapy and broad-spectrum antibiotics are commonly used [3]. Although some studies indicate that cyclosporine A [4, 5, 6], polymyxin B-immobilized fiber column hemoperfusion [7, 8] and anticoagulant agents [9] might be beneficial for treating AE-IPF, mortality remains high. Therefore, a new therapy that improves the prognosis for AE-IPF is urgently needed.

Thrombomodulin is a thrombin receptor on the endothelial cell surface and has an important role in regulating intravascular coagulation [10]. Recombinant human soluble thrombomodulin (rhTM) is composed of the active extracellular domain of thrombomodulin. rhTM forms a reversible complex with thrombin to convert plasma protein C into activated protein C, which inactivates coagulant factors and the proinflammatory effects of thrombin. Moreover, rhTM directly binds and sequesters high-mobility group box 1 (HMGB-1), leading to suppression of inflammation [11]. Previous studies reported coagulation abnormalities in IPF [12, 13] and elevated HMGB-1 levels in the bronchoalveolar lavage fluid from AE-IPF patients [14]; thus, control of the coagulation system and suppression of inflammation by inhibiting HMGB-1 might improve the outcomes of patients with AE-IPF. We examined the clinical effectiveness of rhTM and evaluated whether it had a beneficial effect on the survival of patients with AE-IPF.

We retrospectively reviewed the medical records of 402 consecutive IPF patients who were admitted to Toho University Omori Medical Center in the period from April 2006 to March 2013. A total of 41 patients who had received a first clinical diagnosis of AE-IPF were included in this study; 16 were treated with rhTM (rhTM group), and 25 were treated without rhTM (control group).

According to the American Thoracic Society/European Respiratory Society/Japanese Respiratory Society/Latin American Thoracic Association guideline [15], IPF is diagnosed on the basis of histologic findings from a lung biopsy, high-resolution computed tomography (HRCT) images, or both. Six patients (15%) underwent surgical lung biopsy before AE-IPF onset, and all specimens showed a UIP pattern. The HRCT images of all the patients were reviewed by 2 pulmonologists (T.I. and S.S.) and 1 chest radiologist (A.K.). In Japan, the classification of IPF disease severity is used to make decisions regarding the subsidization of medical care [16]. This classification scheme, which was used in this study, was as follows: stage I (PaO₂ ≥80 mm Hg at rest), stage II (PaO₂ 70-79 mm Hg at rest), stage III (PaO₂ 60-69 mm Hg at rest) and stage IV (PaO₂ <60 mm Hg at rest). Patients with stage II or stage III disease who experience desaturation during a 6-min walk test are classified as stage III or IV, respectively. Disease stage, pulmonary function and dyspnea scale were assessed while IPF was chronic and stable before AE-IPF onset.

AE-IPF was defined based on criteria proposed by Collard et al. [1] and the guideline of the JRS [16], with slight modifications, is as follows: (1) a previous or current diagnosis of IPF, (2) unexplained worsening or development of dyspnea in the past 30 days, (3) an HRCT scan showing new bilateral ground-glass opacities and/or consolidation superimposed on a background reticular or honeycomb pattern, (4) no evidence of pulmonary infection on bronchoalveolar lavage, endotracheal aspiration or sputum culture and negative results on blood tests for other potentially infectious pathogens (e.g. Pneumocystis jiroveci, cytomegalovirus) and (5) exclusion of left heart failure, pulmonary embolism and other possible causes of acute lung injury. Infectious disease was excluded by examination of several microbiological samples. The results were negative for sputum culture of bacteria, urinary antigen tests for Streptococcus pneumonia and Legionella pneumophila and serologic studies for the viruses Chlamydia pneumonia and Mycoplasma pneumonia. Left heart failure and pulmonary embolism were excluded by echocardiography, tests of brain natriuretic peptide (BNP) and D-dimer, and, if necessary, by enhanced CT. Using the classification of CT patterns described by Akira et al. [17], we classified the CT pattern of all patients at the onset of AE-IPF as either diffuse, peripheral or multifocal.

AE-IPF was treated with high-dose CS pulse therapy (methylprednisolone 1,000 mg/day for 3 days) in all patients. The CS dose was tapered after pulse therapy (0.5-1.0 mg/kg/day) and in almost all patients, CS therapy was combined with cyclosporine A (2.5 mg/kg/day). Before November 2011, some patients in the control group received low-molecular-weight heparin (LMWH), 75 IU/kg/day for 14 days, based on the findings of a report by Kubo et al. [9]. Since November 2011, rhTM replaced LMWH and was administered at a dose of 0.06 mg/kg/day for the first 6 days, in combination with CS therapy. Outcomes after the onset of the first AE-IPF episode and adverse events after the start of treatment were compared between the rhTM group and the control group. We also compared outcomes between the rhTM group and patients in the control group treated with LMWH.

All clinical and laboratory data were collected from patients' medical records. Continuous variables are expressed as mean ± SD unless otherwise stated and were compared using the Mann-Whitney U test. Categorical variables were compared using the χ² test. Survival was investigated by using the Kaplan-Meier method, and differences were assessed by the log-rank test. Cox proportional hazards regression analysis was used to identify variables that were significant predictors of survival. A p value <0.05 was deemed statistically significant. All statistical analyses were performed using SPSS version 11.0 (SPSS Inc., Chicago, Ill., USA).

This study was approved by the Institutional Review Board of Toho University Omori Medical Center (project approval No. 23-168). All patients or their families provided written informed consent, and medical records were reviewed with the approval of the Institutional Review Board.

We identified 41 patients (36 men and 5 women) who had been treated for AE-IPF. The median observation period, from the first consultation at our center, was 12 months (range 1-143 months). Thirty-six patients (88%) had a smoking history. Nine patients (22%) had a pathologic diagnosis of UIP, which was revealed by analysis of a surgical lung biopsy specimen obtained 0-36 months before AE-IPF onset (n = 6) or by autopsy (n = 3). Histologic findings in all autopsy cases showed diffuse alveolar damage superimposed on the UIP pattern. Twenty-two patients (54%) received treatments such as CS or antifibrotic agents before AE-IPF onset.

Table 1 shows the clinical characteristics of the 2 patient groups. D-dimer level significantly differed between groups. There were no significant differences between groups in other baseline characteristics including pulmonary function test results before AE-IPF onset, PaO₂/FiO₂ ratio and serological markers at AE-IPF onset. The interval from the last pulmonary function test to AE-IPF onset was short in both groups (2.3 ± 1.7 vs. 3.2 ± 2.7 months in the rhTM and control groups). HRCT at AE-IPF onset showed diffuse ground-glass opacities superimposed on preexisting subpleural fibrosis. A few patients in both groups showed consolidation combined with ground-glass opacities. The diffuse CT pattern was the most frequent pattern in both groups at AE-IPF onset. There were no significant differences between groups in the characteristics of the radiologic images.

Treatments for AE-IPF are shown in table 2. All patients received high-dose CS pulse therapy followed by maintenance treatment with a tapered dose of CS, and most patients in both groups also received cyclosporine A. There were no significant treatment differences between groups, except for the LMWH treatment: 11/25 patients (44%) in the control group received LMWH at the start of treatment.

During the observation period, 29/41 patients (71%) died. Twenty-four patients (84%) died from respiratory failure caused by AE-IPF (n = 21) or chronic disease progression of IPF (n = 3). In addition, 1 (3%) died of P. jiroveci pneumonia, 1 (3%) from cytomegalovirus infection, 1 (3%) of stomach cancer and 2 (7%) due to unknown causes. Figure 1 shows the survival curves for the rhTM and control groups. Survival at 3 months was significantly better in the rhTM group than in the control group (survival rate 69 vs. 40%, p = 0.048). Overall survival after AE-IPF onset was also significantly better in the rhTM group (median survival time: 165 vs. 53 days, p = 0.031; fig. 2). Figure 3 shows the survival curves at 3 months for the rhTM group and the LMWH-treated patients in the control group (n = 11); there was no significant difference between groups (survival rate: 69 vs. 46%, respectively; p = 0.17). The univariate Cox proportional hazard regression model showed that the factors predicting survival were lactate dehydrogenase (LDH; hazard ratio 1.003, 95% confidence interval 1.000-1.006; p = 0.020) and rhTM treatment (hazard ratio 0.446, 95% confidence interval 0.210-0.948; p = 0.036; table 3). Age, gender, other serological markers (including D-dimer), CT pattern and other AE-IPF treatments besides rhTM were not factors for prognosis in this study.

Mild hemoptisis and hematuria developed in 1 patient in the rhTM group, on the day after rhTM administration. These symptoms improved within a few days, without stopping the rhTM treatment. Severe bleeding did not develop in any patient in either group.

This is the first retrospective study of rhTM treatment for AE-IPF. Survival at 3 months was significantly better in the rhTM group than in the control group, and rhTM treatment predicted survival in a univariate Cox proportional hazard regression model. No severe bleeding events developed with rhTM treatment. Our findings suggest that rhTM treatment improves outcomes in AE-IPF and is safe under conditions of routine clinical monitoring.

Several studies reported an association between the presence of coagulation disorders and AE-IPF. Kotani et al. [12]found that procoagulant tissue factor levels in bronchoalveolar lavage fluid were higher in IPF patients than in normal subjects and that these levels correlated with disease activity. Collard et al. [13]reported significant elevations in plasma biomarkers of endothelial cell injury and coagulation in patients with AE-IPF. In addition, serum thrombomodulin level was a significant prognostic marker. HMGB-1 may be a late inflammatory mediator, and elevated HMGB-1 concentrations were observed in patients with sepsis [18] and acute lung injury [19]. Ebina et al. [14]found that HMGB-1 level was elevated in bronchoalveolar lavage fluid from AE-IPF patients. rhTM directly inhibits HMGB-1, leading to an anti-inflammatory effect. It also binds thrombin and inactivates the coagulation system. Therefore, rhTM treatment is likely to have a beneficial effect on AE-IPF patients.

Previous reports showed that rhTM was effective in treating disseminated intravascular coagulation and sepsis [20, 21, 22, 23, 24]. Kato et al. [21]reported that disseminated intravascular coagulation score on day 7 was significantly lower in an rhTM treatment group than in a control group. In addition, Saito et al. [20]noted a better rate of DIC improvement when compared with a heparin-treated group. Little is known about the effectiveness of anticoagulant agents in IPF. Kubo et al. [9]reported that anticoagulant agents, including heparin and warfarin, resulted in a significant survival benefit for IPF patients when coadministered with CS when compared with CS treatment only. In contrast, in the ACE-IPF trial [25], warfarin use by patients with progressive-stable IPF was associated with increased mortality when compared with placebo, which suggests that warfarin should not be used to treat chronic IPF. However, there are no data on rhTM therapy for AE-IPF.

Previous studies reported that conventional CS treatment resulted in a 3-month survival rate of 30-40% after AE-IPF onset [3, 26]. In this study, the control group receiving conventional treatment had a similar survival rate. Survival rate was significantly better in the rhTM group than in the control group, and rhTM treatment was a significant prognostic factor for survival in the univariate analysis, which suggests that rhTM treatment has a beneficial effect on survival in AE-IPF.

D-dimer is a final product of cross-linked fibrin degradation and is released into circulation during endogenous fibrinolysis. It is believed to be a useful marker of abnormal coagulation balance [27] and may be influenced by increased intra-alveolar fibrin deposition [9]. In this study, the plasma D-dimer level was higher in the control group at AE-IPF onset. Although it is possible that coagulation imbalance was worse among patients in the control group, D-dimer level was not a prognostic factor in the univariate analysis. Previous reports found that prognostic factors for AE-IPF were a diffuse distribution on HRCT images, LDH level, C-reactive protein level after AE-IPF onset, greater impairment of pulmonary function and longer duration between admission and the start of AE-IPF treatment [17, 25, 28]. However, to our knowledge, no previous study reported that D-dimer level was a predictive factor for prognosis or severity in patients with AE-IPF. The prognostic importance of D-dimer level in AE-IPF requires future study.

Increased risk of bleeding is the greatest concern with rhTM administration. At clinical blood concentrations, the mechanism of action for rhTM is such that thrombin generation is suppressed via activated protein C, without direct inhibition of thrombin activity. rhTM has a greater safety margin than other anticoagulant agents such as heparin [29], and rhTM dosage can be set so as to provide potent anticoagulant activity while minimizing bleeding, thus allowing for stable outcomes. In this study, we found that the incidence of bleeding complications was not increased by rhTM administration, which suggests that rhTM therapy can be safely used during routine clinical monitoring.

This study has several limitations. First, it was a retrospective study at a single center and we were unable to conduct multivariate analysis because of the relatively small sample size. Therefore, larger-scale prospective studies are needed in order to confirm our results. Second, some of the patients in the control group were treated with LMWH, as recommended by Kubo et al. [9]. There was no significant difference in survival between the rhTM group and LMWH-treated patients. Future studies should compare the effectiveness of rhTM and LMWH for treating AE-IPF in a larger sample. Third, the D-dimer level differed significantly in the 2 groups. It is possible that coagulation imbalance was worse among patients in the control group; however; to our knowledge no study has reported that D-dimer level was a predictive factor for prognosis or severity in patients with AE-IPF. Fourth, not all patients in our series underwent bronchoalveolar lavage and enhanced CT in order to rule out the possibility that respiratory failure was caused by infection or pulmonary embolism. In all the cases where bronchoscopy was not performed due to severe respiratory failure, infection was ruled out by using less invasive procedures such as sputum and/or blood culture, urinary antigen tests and serologic analysis. In patients who did not undergo enhanced CT due to renal dysfunction or allergy to contrast medium, pulmonary embolism was excluded by echocardiography, serum BNP level and D-dimer level. Fifth, due to severe respiratory failure, the PaO₂ in ambient air was not examined in some patients. This meant that when the PaO₂/FiO₂ ratio at the onset of AE-IPF was compared, it did not differ between groups. Finally, AE-IPF can be fatal and requires various types of supportive care. Since the quality of general supportive care might improve the outcome, a similarity in baseline features in different groups does not guarantee that the outcome would be the same if the therapy were really ineffective.

In conclusion, rhTM can be used safely and may have a significant beneficial effect on mortality in patients with AE-IPF. AE-IPF is a fatal condition and therapeutic guidelines have not yet been established. rhTM therapy in combination with conventional treatment appears promising for improving the poor prognosis of AE-IPF. However, this study provides only preliminary data with several limitations and the effectiveness of rhTM on AE-IPF could not be established. Therefore, large placebo-controlled randomized trials are required in order to confirm our findings.

This study was supported by a grant from the Ministry of Health, Labour and Welfare of Japan awarded to the Study Group on Diffuse Lung Disease, Scientific Research/Research on Intractable Diseases.