Results
Baseline Characteristics
Baseline characteristics of patients enrolled in the study are demonstrated in Table 1. In total, 40 consecutive patients with CPFE and 60 patients with IPF were retrospectively identified from the archive records of three different hospitals, during the period of time between September 2004 and August 2010. Three patients with CPFE diagnosed with MPA based on clinical evidence (hemoptysis, hematuria), serum immunologic profile and renal biopsy showing pauci-immune necrotizing glomerulonephritis, were treated with pulses of methyprednisolone (1gr for 3 consecutive days) and cyclophosphamide (1gr for one day) and then, as maintenance treatment, with oral cyclophosphamide (60 mgr/day) and high doses of oral prednisolone (60 mg/day) that were gradually tapered, for a total period of 7 months (n = 3), as previously described. One patient died due to respiratory and renal failure. The remaining 4 cases with CPFE presenting with positive p-ANCA and asymptomatic microscopic hematuria, with normal renal function and mild gas exchange impairment, evidence suggesting but not establishing the diagnosis of MPA, were switched from pirfenidone treatment to a more anti-inflammatory therapeutic regimen comprising of low doses of prednisolone (20 mgr/day), azathioprine (2 mg/kr/day) and high doses of NAC (1800 mgr/day).
Regarding pulmonary functional profile at the time of disease diagnosis, CPFE patients enrolled in our study exhibited a mild restrictive pattern with relatively preserved lung volumes and disproportionally impaired gas exchange and exercise capacity as indicated by DLCO and 6-minute walking distance (6MWD) values (Table 1). Compared to patients with IPF, patients with CPFE demonstrated significantly higher FVC%pred and TLC%pred and lower DLCO, MMEF25/75%pred and 6MWD values (Table 1). Finally, patients with CPFE presented with a statistically significant elevated levels of systolic pulmonary artery pressure (sPAP) levels (37.3 ± 14.5 mmHg) compared to IPF patients (32.1 ± 10.9 mmHg) (p < 0.05), as assessed by cardiac ultrasonography (Table 2).
HRCT findings showed that the mean global extent of emphysema was 15.7% (range: 5%-56%) and the mean global extent of interstitial lung disease was 42.2% (range: 6%-80%).
Serum Immunologic Profile
All patients enrolled in the study were subjected to a complete serum immunologic profile. Strikingly there was a statistically significant increased number of CPFE patients presenting with elevated (>1/160) serum titers of ANA (n = 17/40, 42.5%) compared to patients with IPF (n = 16/60, 26.6%), (p < 0.05) (Table 3). Among these patients 15/17 (88%) exhibited positive ANA profile at the time of CPFE diagnosis, therefore were naïve of treatment, whereas in the remaining 2/17 (12%) ANA serum titers became positive after a mean period of 12 months following disease diagnosis. With regards to IPF subjects, 9/16 (56.2%) reported positive ANA profile at the time of diagnosis whereas the remaining 7/16 (43.8%) became positive after a mean follow-up period of 18 months. Furthermore, one of the most intriguing findings of our study was the identification of a higher proportion of CPFE ANA + patients that also exhibited positive ANCAs (n = 7/40, 17.5%) against MPO compared to IPF patients without emphysema (n = 0) (p < 0.05). Presence of autoantibodies was estimated by both qualitative (immunofluorescence) and quantitative (ELISA, mean levels = 15,6 elisa units –EU, normal range <9.0 EU) methods (Table 2). In 6 out of 7 patients ANA profile was positive and p-ANCA profile was negative at the time of CPFE diagnosis and became positive after several months (mean period of delay = 20 months) during either disease exacerbation (respiratory and renal failure) (n = 2) or in the setting of routine follow-up (n = 4). In one patient both ANA and p-ANCA concentrations were found elevated at the time when pulmonary and renal involvement were diagnosed. In 3 out of 7 CPFE patients diagnosis of MPA was firmly established by renal biopsy showing pauci-immune necrotizing glomerulonephritis while microscopic urine analysis revealed red blood cell renal casts. The latter group of patients exhibited the most aggressive clinical course comprising of respiratory and renal failure derivative of diffuse alveolar hemorrhage and necrotizing renal vasculitis. The remaining 4 CPFE patients with positive ANA and p-ANCA profile were characterized by an asymptomatic microscopic hematuria, with normal renal function and mild gas exchange impairment (Table 4).
There were no differences between the three groups of CPFE patients based on their immunologic profile (ANA + ANCA+, ANA + ANCA-, ANA-ANCA-) with regards to baseline functional (FEV1, FVC, DLCO, sPAP) and radiological (HRCT pattern) parameters (Table 3). Nevertheless, Kaplan Meier analysis revealed a statistically significant improved survival of patients with CPFE and positive ANA profile compared to those with negative ANA titers (median survival 51 months-range 12–96 months vs. 38 months-range 16–61 months, p = 0.052, respectively) (Figure 1a). There were no survival differences in patients with IPF with and without positive immunologic profile (data not shown), while renal function and serologic profile returned to normal levels.
(Enlarge Image)
Figure 1.
Kaplan Meier survival and spearman's correlation analysis n patients with CPFE with positive and negative immunologic profile. a. Kaplan-Meier survival curve revealed a marginal statistically significant improved survival (p = 0.052) of patients with CPFE and positive ANA profile (median survival of 51 months, range 12 – 96 months) compared to those with negative immunologic profile (38 months, range 16 – 61 months) b. Spearman's correlation analysis revealed an almost linear positive association between the number of CD20+ cells/mm and median survival of patients with CPFE.
Three CPFE and 6 IPF patients with symptoms of morning stiffness and mild arthralgia exhibited marginally positive rheumatoid factor (RF) levels (>20) while one from the CPFE and two from the IPF group of patients presented with positive ENA panel as assessed by elevated circulating antibodies against topoisomerase (anti-scl70) indicative of systemic sclerosis one and two years, respectively, after the initial diagnosis of CPFE and IPF. Finally all patients exhibited negative myositis panel.
Histopathology Profile
Evaluation of the histopathology profile was performed by two independent pathologists in tissue microarray blocks derived from lung biopsy samples of 15 patients with CPFE, 28 patients with IPF and 15 controls. A pattern of definite UIP was revealed in 10 CPFE and 18 IPF cases and a probable UIP pattern in 5 CPFE and 10 IPF cases. Based on evidence arising from the investigation of the serum immunologic profile, we performed immunohistochemical analysis using CD20 antibody. The number of CD20 positive cells/mm was significantly higher in CPFE ANA + (n = 7) (49.6 ± 7.1) and IPF ANA + (n = 9) (47,1 ± 6.4) compared to CPFE ANA- (n = 8) (6.2 ± 3.6), IPF ANA- (n = 19) (5.9 ± 1.6) and control lung samples (3.2 ± 1.1) (p < 0.05) (Table 3) . There was no difference in the number of CD20+ cells between CPFE ANA + and IPF ANA + patients. Further subgroup analysis revealed that CPFE ANCA + patients with available lung biopsy samples (n = 5) also exhibited statistically significant increased CD20+ cells infiltration (51.6 ± 6.9) compared to CPFE ANCA- patients (n = 7) (7.1 ± 2.2) (p < 0.05). To further extend our previous observation that presence of positive autoimmune profile may be a positive prognosticator we performed Spearman's correlation and clearly demonstrated an almost linear positive association between the number of CD20+ cells/mm and median survival in patients with CPFE (Figure 1b). There were no similar correlations in patients with IPF (data not shown).
Finally and most intriguingly, as depicted in Figure 2, CD20+ B cells forming lymphoid follicles were found within the fibrotic interstitium in areas adjacent to fibroblastic foci in both CPFE ANA + (A, B) and CPFE ANCA + (C, D) lung samples whereas CD20+ cells were almost absent in CPFE ANA-ANCA- (E, F) as well as in control lung samples (G, H), supporting the presence of auto-antibody producing cells. There was no difference in CD20 + cells/mm between CPFE ANA + (49.6 ± 7.1) and CPFE ANCA + (51.6 ± 6.9) patients.
(Enlarge Image)
Figure 2.
Presence of CD20+ B cells within the fibrotic interstitium in patients with CPFE. Representative immunohistochemistry with a CD20 antibody on lung paraffin sections from patients with CPFE with a UIP histopathologic pattern and controls. As illustrated clusters of CD20+ B cells forming lymphoid follicles were found within the fibrotic interstitium in areas adjacent to fibroblastic foci in CPFE patients with ANA + profile (a, b) as well as CPFE patients with ANCA + profile (c, c), whereas CD20+ cells were almost absent in biopsy samples from patients with CPFE and negative ANCA and ANA profile (e, f), as well as in control lung samples (g, h). Original magnification : ×200 (a, c, e and g), × 400 (b, d, f and h).