© 2008 American Thoracic Society doi: 10.1164/rccm.200802-304ED
Scleroderma Lung Disease"If You Don't Know Where You Are Going, Any Road Will Take You There"
Medical University of South Carolina
University of Michigan If Lewis Carroll (quoted in the title) were alive today, he may well have considered a career in scleroderma research. Systemic sclerosis (SSc; scleroderma) remains without a clearly defined pathogenesis and without an effective overarching approach to therapy. Despite the fact that survival from scleroderma appears to be improving, progress has come slowly through development of specific organ–based strategies such as angiotensin-converting enzyme inhibitors for scleroderma renal crisis. Unfortunately, lung involvement has emerged as the leading cause of death with interstitial lung disease (ILD) and pulmonary arterial hypertension of nearly equal impact (1). Two recent prospective, randomized, controlled trials suggest that cyclophosphamide may slow loss of pulmonary function in patients with scleroderma ILD (2, 3). The Scleroderma Lung Study (SLS) employed daily oral cyclophosphamide for 1 year in patients with mild to moderate ILD and with evidence of "active" pulmonary inflammation—abnormal bronchoalveolar lavage (BAL) and/or ground-glass opacities on high-resolution computerized tomography (HRCT). Oral cyclophosphamide was associated with a 2.53% reduction of loss of FVC over 1 year along with a modest reduction in skin involvement, improved quality of life, and improved dyspnea (2). In a smaller British study, monthly intravenous cyclophosphamide was associated with a 4.19% reduction of loss in FVC, although these results failed to achieve statistical significance (3). In many ways, the findings from the two clinical trials raise more questions than answers. The treatment effect for FVC and for skin involvement was within the intrapatient variability for both measures and one might fairly consider if this represents a "minimally important difference." The beneficial effect of cyclophosphamide was not durable, with most differences between drug and placebo essentially absent at 24 months of follow-up (4). Cyclophosphamide is rational therapy. Abundant evidence supports that inflammation is associated with and perhaps precedes parenchymal fibrosis (5, 6). Yet, in the SLS, findings on BAL were predictive neither of clinical course nor of response to therapy (2, 7). Similarly, ground-glass opacity on HRCT offered little predictive information (2, 7). It has been hypothesized that these trials may have underestimated therapeutic effect in that subjects with active progressive disease were underrepresented (8). Indeed, in the SLS, only 13% of subjects had loss of FVC of 10% or greater, with no difference between cyclophosphamide and placebo in forestalling this level of decline. The question remains as to who with SSc might benefit from therapy. If BAL and HRCT findings of ground glass offer little predictive value in clinical decision making, and if the majority of patients are relatively stable on conservative therapy, it might well be argued that cyclophosphamide, even though supported by evidence, cannot be generally recommended as appropriate therapy. In this respect, the article by Goh and colleagues (9) in this issue of the Journal (pp. 1248–1254) offers a step forward. In a robust analysis of 215 patients, extent of disease by HRCT was shown to be predictive of decline in FVC, diffusing capacity, and survival (9). By classifying subsets of SSc ILD patients with "limited" and "extensive" ILD, they were able to define a group of more advanced patients most likely to progress. Furthermore, the HRCT extent used for the stratification was proven reproducible between ILD extent scores of less than 10% and greater than 30% lung involvement. By adding spirometric data of an FVC of less than 70% predicted to refine the breakpoints between the two subsets (10–30% disease extent), there were few misclassifications. Since there are currently no robust biomarkers that define the progressive patient, these new criteria hold the hope of improving clinical trial design in SSc. These data are congruent with the SLS results wherein severity of baseline fibrosis had a strong influence on response to treatment (2). Will this road get us there? Many trials in idiopathic pulmonary fibrosis (IPF) have shown that individual baseline physiology or HRCT parameters are associated with worse survival (10). In addition, a 6-month decline in FVC or DLCO provides further prognostic information (11, 12). But good stratification systems have not been accepted in this disease (13) because mortality often occurs quickly. Why do we need a surrogate endpoint when we can use the "gold standard" of mortality? SSc ILD is most often slowly progressive. A two-tiered staging system for ILD will allow focused intervention on individuals more likely to progress. Rather than spend years trying to refine the prediction model by adding more risk factors for progression, the current suggestion by Goh and colleagues should be adopted. Prospective refinement of this clinically useful model over time is preferable to following the path of IPF in which no stratification system is accepted. A good first step would be to reanalyze the SLS data using the proposed model. The strategy in scleroderma lung disease when disease extent is mild on HRCT is observation. To determine who deserves procrastination and careful follow-up (limited ILD) and who requires therapy (extensive ILD) is a timely goal. The staging system offered by Goh and colleagues (9) offers the promise of more robust clinical trial design and clarity to guide future SSc ILD management. FOOTNOTES Conflict of Interest Statement: C.S. has no financial relationship with a commercial entity that has an interest in the subject of this manuscript. J.R.S. served on Actelion's International Scientific Advisory Board and received $3,000 per year from 2005–2007; he was International Principal Investigator on the BUILD-2 study investigating bosentan in scleroderma ILD but received no direct funding. REFERENCES
Related articles in AJRCCM:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
