AFOP is a histopathologic diagnosis that was first reported in 2002 by Beasley et al. in a case series involving 17 patients with acute respiratory failure[1] and was defined as a subgroup of idiopathic interstitial lung disease [2] in 2013 in the American Thoracic Society/European Respiratory Society statement. The pathology characterized by fibrin deposition in pulmonary alveoli and organization of loose connective tissue. The other characteristic feature is the absence of Classical hyaline membranes of diffuse alveolar damage and no eosinophil infiltration or granuloma formation.
We performed a literature review of all reports of patients who were diagnosed AFOP over the period from 2002 to 2019, by conducting thorough search in PubMed and Web of Science databases, and searching for the reference lists of relevant articles for further reports as well. The search was limited to publications in English. By Dec 2019, there have been 85 published reports of 160 patients in the literature. However, 4 studies including 12 patients were removed for lacking sufficient clinical data of each case, including 6 SARS patients and 4 patients post lung transplantation reported by Hwang et al and Bierach
et al respectively. Finally, a total of 81 eligible reports[3-26] with non-overlapped 148 evaluable cases qualified for the literature review and 150 cases (including the patients in the present study) served as the study population of this analysis.
There was no clear gender difference and AFOP most commonly happened in patients in their fifth to sixth decade of life. Of the 150 cases of AFOP reviewed, 65 were male, 63 were female and the other 22 (by Paraskeva et al[26] )were not reported; the mean age was 54.3±15.8 years (range, 38 days‐84 years) (Table 1). There were about 78% patients between 40 and 70 years old at the time of diagnosis, while there were 3 reports in children (one infant). In our case reports, both were female in their fifties, consistent with the average reported age. AFOP have been reported mainly in USA, Europe and the Asia area. Race was not always included in publications; however, when it was reported, patients usually were Caucasian(60 cases)and Asian(37). There were fewer Hispanic patients(10), and only 4 black patients were described, while race was not specified in 43 cases.
The etiology of AFOP can either be idiopathic or associated with a wide range of medical conditions or risk factors, such as lung transplantation, cancer, medication reaction and so on. In the literature review, 50 cases had no underlying cause or association, diagnosed as idiopathic AFOP, while the other 100 cases were identified as secondary AFOP (Table 2). In addition, 17 (11.3%) patients had no significant medical history and were apparently healthy prior to the present illness (Table 2). It should be noted that there were two cases caused by immune checkpoint inhibitor (ICI) (Nivorumab and pembrolizumab, respectively for specific)[8, 21], which is one of the most popular and effective treatments for metastatic melanoma and advanced non-small cell lung cancer in recent years. AFOP should be taken into account when considering ICI‐related pneumonitis for PD1/PD‐L1 inhibitors. Smoking appeared not to be a risk factor for only 22(38%) had a history of smoking, of the 58 patients with available data. The smoking ratio seemed to be higher in the secondary-AFOP populations than in the idiopathic-AFOP group, with 47%( 14/30) to 29%(8/28).
Clinical manifestations of AFOP appeared to be nonspecific. Most common symptoms reported included dyspnea in 108 (72%), cough in 106 (71%) (87 had a non‐productive cough), fever in 64(43%), and chest pain in 24 (16%) (Table 1). Other less common symptoms included hemoptysis, fatigue, chills, night sweat and weight loss. One patient was asymptomatic. Fever was considered to be more common in AFOP when compared with other forms of interstitial pneumonia[27]. The disease progression was acute in 41 patients, subacute in 62 patients, while the remaining were unspecific.
Most patients underwent chest CT scan, except 2 patients with acute process only had X- rays and radiologic images were not specified in 3 cases. The most common CT findings were bilateral patchy infiltrate and diffuse consolidation, often with basal or peripheral distribution. One hundred and sixteen patients presented a bilateral distribution, while 18 cases were limited to unilateral lung (the ratio of right to left was 2 to 1). Lung lesion was localized or greater in lower lobes in 29 cases, while 11cases reported mainly in the upper lobes. The most common imaging pattern was consolidation in 79 patients, of which there were 28 cases together with ground-glass opacity(GGO); GGO in 61 patients and nodules in 29 cases (Table 1). Compared with patients of secondary-AFOP, patients of idiopathic -AFOP showed more imaging finding with consolidation and less proportion with GGO. Air bronchogram was often reported with consolidation. Other less common radiographic findings included traction bronchiectasis, mild or medium hilar and mediastinal adenopathy and halo sign. Pleural involvement seemed not to be common presentation, for only 7 cases were with mild bilateral effusion, 3 with slight right pleural effusion and only 2 patients had pneumothorax. In addition, the images present as interstitial pneumonia in 11 cases and showed the lesion as a solitary mass suggestive of a primary lung tumor in 2 cases. All of these atypical images were seen in the Secondary-AFOP.
The nonspecific clinical features result in misdiagnosis and a delay in the diagnosis of AFOP. At the time of initial diagnosis, most cases were misdiagnosed as lung infection, while a few cases were misdiagnosed as lung cancer. Of the cases caused by non-infectious factors, 81 cases were empirically treated with antibiotics, the vast majority with broad-spectrum antibiotics and some with antifungal therapy, when 60 cases were not specific. Beasley and his colleagues[1] described a mean time from onset of symptoms to diagnosis of 19 days, and Gomes et al[4] reported a mean time of 43.9 days.
A lung biopsy is required for a definitive diagnosis of AFOP, including examinations for microbiology to exclude infections. The lung lesion tissue was procured by surgical lung biopsy (mainly by video-assisted thoracic surgery, VATS) in 60 patients, transbronchial lung biopsy in 43 patients, image-guided (CT- or ultrasound-guided) percutaneous lung biopsy in 31 patients, autopsy in 13 patients and unspecified lung biopsy techniques in 3 patients. Of the cases reviewed, 28 patients underwent BAL before pathological diagnosis and no conclusive findings were found. Both of our cases’ BAL culture and transbronchial biopsy were inconclusive and the diagnosis was confirmed by percutaneous lung biopsy. It is best diagnosed with open-lung biopsy (or video-assisted thoracic surgery biopsy), given the risk of missing areas of hyaline membranes of diffuse alveolar damage. However, we suggest that the method of lung biopsy should be based on the actual clinical situation, including the location of lung lesions and the patient’s tolerance to the operations. The differential diagnosis should include bronchiolitis obliterans organizing pneumonia, eosinophil pneumonia, and diffuse alveolar damage.
Regarding therapy for AFOP, there are still no standard treatment recommendations. The most common therapy in the literature for AFOP was corticosteroids, with 132 out of 150 patients( 88%). There is no consensus on the dose and duration of corticosteroids and the therapy should be individualized, according to the etiology, clinical course and response. The initial dosage of steroids was reported in 55 cases, varying from prednisone 0.5 mg/kg/day (or equivalent) to methylprednisolone 1000 mg/day, with 28 (51%) cases of prednisone 0.5-1.5 mg/kg/day. A pulse therapy of corticosteroids (prednisone 500mg/day to methylprednisolone 1000mg/day) was administered in 12 cases. Relapse may occur during steroid tapers and symptoms mostly could relieve again when higher doses were resumed. In our two cases, steroid was prescribed as soon as the diagnosis was established and relapse happened when steroid reduced.
Other drugs reported in the literature were antibiotics (especially for those caused by infection), immunosuppresants, etanercept, indomethacin and immunoglobulin, with varying results. Immunosuppressive agents such as cyclophosphamide, mycophenolate mofetil, cyclosporine, azathioprine, tacrolimus and so on, had been tried in 17 patients, all combined with corticosteroids. Patients secondary to autoimmune diseases may benefit most from immunosuppressant combined with steroids. GL Simmons et al[25] reported the first case AFOP after HSCT in 2016 successfully treated with etanercept, a TNF inhibitor, and Zhou et al reported that lowdose indomethacin combined with methylprednisolone was a new choice of treatment for AFOP after
a surgical resection of rectal adenocarcinoma[28]. AFOP associated of anti-EJ autoantibodies with necrotizing myopathy was recently reported with good response to intravenous methylprednisolone and immunoglobulin[23].
Other therapeutic interventions included drug discontinuation, surgical resection, and organ transplantation. For patients suspected to be caused by drugs or environmental/occupational exposures, withdrawal is the most fundamental treatment. In 13 cases caused by a medication, treatment with corticosteroids and discontinuation of the drug mostly resulted in improvement and resolve, except 2 patients died even though they were treated combined with immunosuppressive agents and 2 cases died of other complication. Lung resection was reported as a successful cure while the disease is relatively localized, by Santos C et al from Portugal[29]. Labarinas S et al described a case of AFOP associated with hepatic failure and very severe aplastic anemia, responding to immunosuppressive therapy and hematopoietic stem cell transplantation[30]. There was one report of bilateral lung re-transplantation with subsequent improvement of symptoms in a patient suffering AFOP after lung transplantation[31] and lung transplantation was also reported last year by Alessio Campisi (Italy) as a successful cure in two cases of idiopathic AFOP[6]. Though AFOP has been described following organ transplantation, organ transplantation may be indicated to treat AFOP refractory to medical treatment[6, 30, 31]. However, further studies are required to find out whether these kind of patients will undergo recurrence of AFOP.
The prognosis of AFOP appears to have two different clinical patterns: one is an acute onset with a fast progression, more likely leading to respiratory failure and death, and the other is a subacute form associated with a less fulminating course and a more favorable outcome. Among the 150 case reports, 59 (39%) patients died (10 were unrelated to AFOP) (Table 1). Of those 49 cases died of AFOP , the progression was acute in 17(81%) cases while 28 cases were not specified. Median time between onset of symptom to death of 22 AFOP patients after lung transplantation reported by Paraskeva et al was 101 days[26]. AFOP after lung transplantation seemed to have relatively poor prognosis with poor response to drugs[26]. Patients with acute presentation or secondary-AFOP had higher mortality rate than their corresponding cases. This might due to the secondary-cases with more or worse underlying diseases and lower responsive rate to corticosteroid. Additionally, we found that the development of dyspnea and GGO on CT scan as well as lack of consolidation in AFOP seemed to be associated with poor prognosis. However, findings from the present study require further evaluation.
In conclusion, this is the largest review to date dealing with AFOP, which includes 150 patients of AFOP, reported over an 18-year period. Given the various etiology, nonspecific clinical presentation and insufficient understanding of pathology, AFOP probably had been under diagnosed or misdiagnosis. Our two cases highlight the importance of being aware of this uncommon pattern of acute lung injury. AFOP should be considered when encountering a suspected pulmonary infection case unresponsive to empirical antibiotic therapy, especially with bilateral, basilar or peripheral-predominant consolidation, with or without GGO or nodular infiltration on CT scan. We highly suggest appropriate lung biopsy and thorough pathologic evaluation for these patients, to establish the diagnosis and to improve the outcome by administering more aggressive therapy. Corticosteroids was recommended to be the most useful therapy currently, and treatment options should depend on the etiology and disease severity.