Abstract
Rationale: Noninfectious pneumonitis is a known class effect of mammalian target of rapamycin (mTOR) inhibitors.
Objectives: To assess the incidence, radiographic patterns, management, and outcome of pneumonitis in patients with advanced renal cell carcinoma receiving everolimus.
Methods: Clinical study data from 416 patients, randomized to receive everolimus versus placebo, were analyzed for adverse events of pneumonitis. Radiographic studies performed every 8 weeks were subject to a prospective, independent, blinded central review for the presence of findings indicative of pneumonitis.
Measurements and Main Results: Of 274 patients receiving everolimus, clinical pneumonitis was suspected for 37 patients (13.5%) (none with placebo). Nine cases (3.3%) were grade 1 (asymptomatic), 18 (6.6%) were grade 2 (not interfering with daily living), and 10 (3.6%) were grade 3 (interfering with daily living or oxygen indicated). No grade 4 (life-threatening) pneumonitis was observed. Of the 10 patients with grade 3 pneumonitis, 5 had baseline radiological evidence of pneumonitis before everolimus therapy. Twenty of the 37 cases (54.0%) were reversible within the follow-up period; resolution followed dose reduction for 20 patients and treatment discontinuation in 10 patients. Corticosteroid therapy was initiated in 16 cases. Dedicated radiological review of available serial radiographic studies (245 patients receiving everolimus and 132 receiving placebo) found a higher percentage of new radiographic findings even in patients without a diagnosis of clinical pneumonitis who were receiving everolimus versus placebo (38.9 vs. 15.2%).
Conclusions: Early recognition, prompt intervention, and a conservative approach are important in managing the risk associated with noninfectious pneumonitis in association with everolimus.
Clinical trial registered with www.clinicaltrials.gov (NCT 00410124).
Inhibitors of the mammalian target of rapamycin (mTOR), including everolimus, are used for the prevention of solid organ transplant rejection and the treatment of cancer. This class of drugs has been associated with the development of pneumonitis. Little information is currently available on the incidence and outcome of pneumonitis in patients with advanced cancer receiving mTOR inhibitors.
Everolimus is associated with a 13.5% incidence of clinically identified pneumonitis in patients with renal carcinoma. Dedicated radiological review of serial computed tomography scans during treatment showed new radiographic findings in a higher percentage of patients without clinical evidence of pneumonitis. Radiographic findings, outcomes, and a management strategy are described. Adoption of a conservative approach (in the absence of infection) of dose reduction or discontinuation with corticosteroid administration as needed in those with clinical pneumonitis is associated with manageable risk.
Noninfectious pneumonitis, apparently a class effect of mTOR inhibitors, is characterized by noninfectious, nonmalignant infiltrates and will be referred to as “pneumonitis” throughout this report. It has been described with sirolimus (rapamycin), temsirolimus, and everolimus in isolated case reports or retrospective series with a small number of patients (4–11). An incidence of 5–15% of clinical pneumonitis has been reported in transplant recipients receiving sirolimus (8, 12–14). Published reports suggest that it is relatively unaggressive and reversible on discontinuation of therapy in many cases, but there can be significant toxicity (9, 10). A similar experience with clinical pneumonitis has been noted with temsirolimus (15, 16). However, a radiological review of routine serial chest computed tomography (CT) scans during treatment showed a higher incidence (36%) of new radiographic abnormalities (5). Many patients were asymptomatic and continued treatment without progressing to more severe stages.
Pneumonitis was initially reported in association with everolimus therapy in two investigator-initiated phase II trials in small numbers of patients with breast cancer and renal cancer (17, 18). An incidence of 18.7–49% was noted. Many of the cases were grade 1/2 in severity (79 and 63% of affected patients, respectively, in these studies), and pneumonitis resolved after treatment suspension, dose reduction, or corticosteroid treatment. A retrospective independent radiological review of everolimus use in a phase II study of patients with advanced lung cancer found a 25% incidence of new radiographic findings consistent with pneumonitis with a probable or possible relationship to the study drug (19).
Herein, we describe those patients who experienced pneumonitis during the course of the RECORD-1 (Renal Cell Cancer Treatment with Oral RAD001 Given Daily) study of the efficacy of everolimus in RCC and report their management and outcomes. In addition, to address the issue of a higher incidence of radiographic findings while on treatment with everolimus without clinically recognized pneumonitis (as was noted with temsirolimus), we report the results of a dedicated radiographic review of all available serial CT scans.
Between November 2006 and October 2007, 416 patients with metastatic RCC after failure of treatment with sunitinib, sorafenib, or both sunitinib and sorafenib were randomly assigned at a 2:1 ratio to receive treatment with everolimus 10 mg once daily or placebo as part of the RECORD-1 study (ClinicalTrials.gov identifier, NCT00410124) (2, 3). To help formulate a definitive approach to address pneumonitis occurring in this study, an advisory board consisting of several leading pulmonologists was convened by Novartis Pharmaceuticals Corporation. Routine screening of all patients via chest X-ray (CXR) and CT scan was proposed. Advice on the management of pneumonitis was provided in the form of a treatment algorithm (Table 1). CT scans and CXRs were performed on all patients on a routine basis (at 8-wk intervals). Pulmonary function tests (PFTs) were performed at baseline.
Grade | Intervention | Investigations | Everolimus Dose Adjustment |
|---|---|---|---|
| 1 | No specific therapy required | CT scan and PFTs.* Repeat chest X-ray/CT scan every two cycles until return to baseline | No change |
| 2 | Symptomatic only. Prescribe corticosteroids if cough is troublesome | CT scan and PFTs.* Repeat each cycle until return to baseline. Consider bronchoscopy | Reduce dose until improvement to grade ≤1; consider interruption if symptoms are troublesome. Discontinue treatment if recovery to grade ≤1 is not evident within 3 wk |
| 3 | Prescribe corticosteroids if infectious etiology is ruled out. Taper as clinically indicated | CT scan and PFTs.* Repeat each cycle until return to baseline. Bronchoscopy required | Interrupt treatment until improvement to grade ≤1. Restart therapy within 2 wk at a reduced dose if clinical benefit is evident |
| 4 | Prescribe corticosteroids if infectious etiology is ruled out. Taper as clinically indicated | CT scan and PFTs.* Repeat each cycle until return to baseline. Bronchoscopy required | Discontinue treatment |
Pneumonitis was considered to have occurred if the investigator coded for the adverse event of pneumonitis during the trial. These cases were assessed by review of the trial clinical database for the presence of this adverse event, associated symptoms, severity, treatment, and outcome. Decisions on the need for additional procedures, such as bronchoscopy to exclude infectious etiologies or other causes, were made by investigators during the study. In addition, a Medical Dictionary for Regulatory Activities (MedDRA) search strategy of the clinical database was followed to check for coding for other pulmonary events that could indicate pneumonitis or pneumonitis-like cases. These included terms such as alveolitis, interstitial lung disease, pneumopathy, and pulmonary toxicity. Pulmonary adverse events forming the basis for this search strategy are listed in the online supplement.
All events were graded in accordance with the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI CTCAE), version 3.0 (20). For pneumonitis, this corresponds to the following: grade 1 (mild)—asymptomatic, radiographic findings only; grade 2 (moderate)—symptomatic, not interfering with activities of daily living; grade 3 (severe)—symptomatic, interfering with activities of daily living or oxygen indicated; grade 4 (life-threatening or disabling)—life-threatening or ventilatory support indicated.
Advice on management was provided to investigators in the form of a treatment algorithm developed by a pulmonary advisory panel and incorporated into the study protocol as detailed in Table 1.
Serial routine CXRs and CT scans were subsequently subjected to an independent, blinded (with treatment and safety results unknown) central review for the presence or absence of new radiographic findings that could be pneumonitis; categorization of the findings if present; and details of its distribution, severity, incidence of associated pleural effusions, and evidence of progression. An incidence rate was calculated for newly occurring or worsening radiological evidence of pneumonitis that was confirmed on a CT scan.
Imaging findings on chest CT scan were classified into four distinct patterns, consistent with the approach proposed by Endo and colleagues (21): pattern A—nonspecific areas of ground-glass attenuation, corresponding to diffuse and faint opacity without loss of lung volume on chest radiography; pattern B—multifocal areas of airspace consolidation (as in, but not diagnostic of, cryptogenic organizing pneumonia or bronchiolitis obliterans organizing pneumonia), corresponding primarily to peripheral consolidation on chest radiography; pattern C—patchy distribution of ground-glass attenuation accompanied by interlobular septal thickening (as in, but not diagnostic of, acute eosinophilic pneumonia), corresponding to patchy or diffuse faint, linear opacities on chest radiography; and pattern D—extensive bilateral ground-glass attenuation or airspace consolidation with traction bronchiectasis (as in, but not diagnostic of, acute interstitial pneumonia), corresponding to diffuse faint opacities or consolidation with lung volume loss on chest radiography.
Progression-free survival curves based on the absence/presence of clinical pneumonitis were estimated by Kaplan-Meier methodology; cohorts of patients with radiographic findings consistent with pneumonitis versus those with no new findings were compared by log-rank test. The percentage of predicted absolute values for PFTs was calculated using the equations of Crapo (22, 23).
A total of 274 patients received treatment with everolimus and 137 received placebo. Demographic characteristics are summarized in Table 2. Close to 80% of patients were male, and approximately one-third had a history of one or more respiratory disorders and/or symptoms at the start of treatment. This included patients with a wide range of pulmonary diseases including chronic obstructive pulmonary disease, asthma, restrictive pulmonary disease, and/or symptoms of dyspnea and cough, among others. Lung metastases were present in approximately three-fourths of the patients. Previous treatment with interferon had been received by 51 and 52% of patients in the everolimus and placebo groups, respectively. Treatment with interleukin-2 had been received by 22 and 24% of these patients, respectively. In more than 90% of patients, these drugs had been given several months before the start of treatment with everolimus.
Everolimus, 10 mg (n = 274)* | Placebo (n = 137)* | |
|---|---|---|
| Median age, yr (range) | 61.0 (27–85) | 60.0 (29–79) |
| Sex, n (%) | ||
| Male | 215 (78.5) | 105 (76.6) |
| Female | 59 (21.5) | 32 (23.4) |
| Underlying respiratory disorders/symptoms, n (%) | 97 (35.4) | 45 (32.8) |
| Respiratory disorders | ||
| Pleural effusion | 12 (4.4) | 4 (2.9) |
| Chronic obstructive pulmonary disease | 8 (2.9) | 0 |
| Pulmonary embolism | 3 (1.1) | 0 |
| Site of most recent metastases, n (%) | ||
| Lung | 214 (78.1) | 108 (78.8) |
| Lymph node | 153 (55.8) | 82 (59.9) |
| Bone | 104 (38.0) | 46 (33.6) |
| Liver | 99 (36.1) | 48 (35.0) |
Pneumonitis was diagnosed by the investigators in 36 patients (13.1%) who received everolimus therapy. No cases of clinical pneumonitis were reported in patients receiving placebo. After a systematic review of all respiratory events reported by the investigators using other search terms for pulmonary toxicity, three additional cases were considered to be representative of pneumonitis but had been identified as pneumopathy (n = 2) and noninfectious pneumopathy (n = 1). In addition, two cases that had been coded as pneumonitis were noted to be of infectious origin and were therefore discounted. This provided a group of 37 patients (13.5%) with clinical pneumonitis—34 patients as per the study coding and three additional patients by this expanded search.
Of the 37 patients with clinical pneumonitis, a total of 19 (51.4%) patients had cough, with grade 3 cough in one (2.7%) patient (Table 3). Dyspnea was present in 16 (43.2%) patients, with 6 (16.2%) having grade 3 symptoms. A total of 12 (32.4%) patients had both cough and dyspnea.
No Clinical Pneumonitis† | ||||
|---|---|---|---|---|
| Clinical Pneumonitis* (n = 37) | New Radiographic Findings (n = 107) | No New Radiographic Findings (n = 92) | ||
| Cough, n (%) | ||||
| Total | 19 (51.4) | 31 (29.0) | 23 (25.0) | |
| Grade 1 | 10 (27.0) | 22 (20.6) | 17 (18.5) | |
| Grade 2 | 8 (21.6) | 9 (8.4) | 5 (5.4) | |
| Grade 3 | 1 (2.7) | 0 (0.0) | 1 (1.1) | |
| Dyspnea, n (%) | ||||
| Total | 16 (43.2) | 22 (20.6) | 15 (16.3) | |
| Grade 1 | 6 (16.2) | 7 (6.5) | 5 (5.4) | |
| Grade 2 | 4 (10.8) | 8 (7.5) | 6 (6.5) | |
| Grade 3 | 6 (16.2) | 7 (6.5) | 3 (3.3) | |
| Grade 4 | 0 (0.0) | 0 (0.0) | 1 (1.1) | |
| Cough and dyspnea, n (%) | ||||
| Total | 12 (32.4) | 5 (4.7) | 10 (10.9) | |
Time to onset for these pulmonary events ranged over several months. Descriptive statistics for the 37 patients with pneumonitis showed a median time to occurrence of 108 days (range, 24–257 d).
For the 37 patients with a clinical diagnosis of pneumonitis by investigators, radiographic findings were independently reviewed to confirm the presence of new pneumonitis on radiographic study and to assess the nature of the abnormalities. In two cases, radiographs were not available for central review, and 29 patients with CT scans and 6 patients with CXRs only were reviewed. Findings consistent with pneumonitis were confirmed in all but two patients. In one case, changes suggestive of pneumonitis were seen on CXR at the time of diagnosis, but a subsequent available CT scan for independent review by the radiologist suggested that findings had likely been related to progression of cancer. In another case, there were baseline radiographic findings; the local investigators noted a worsening of findings at the time of suspected pneumonitis, but the independent radiologist did not concur with the difference.
For analysis of radiographic patterns, only those with serial CTs throughout the course of study were analyzed. In this subgroup of 25 patients, various radiographic patterns were seen. Furthermore, the presentation changed over time, with 46.7% of patients exhibiting pattern A, 33.3% pattern B, 36.7% pattern C, and 36.7% pattern D (at any time). When reanalyzed in terms of the last pattern retained during the observation period, pattern A was evident for 36.7% of patients, pattern B for 26.7%, pattern C for 10.0%, and pattern D for 26.7%. Diagnoses of pneumonitis were associated with pleural effusion in only two cases—a mild, unilateral effusion was seen in one patient with a grade 2 event and mild, bilateral effusions in one patient with grade 3 pneumonitis.
An exploratory analysis was conducted to determine whether any correlation was evident between CT and CXR data in diagnosing pulmonary events. In 22 patients, both a CXR and CT were available at the same time period during pneumonitis. Positive confirmation was obtained on both CT and CXR for 14 patients but by CT only in 8 patients. The degree of correlation improved with increasing severity (grading) of pneumonitis (50.0% for grade 1 [mild]; 66.7% for grade 2 [moderate]; and 71.4% for grade 3 [severe]).
The maximal NCI CTCAE grading (severity) of pneumonitis and related events is presented in Table 4 for all cases identified as pneumonitis. Grade 1 and 2 cases were present in 3.3 and 6.6% of patients, respectively, with 3.6% having grade 3. There were two deaths in patients with grade 3 toxicity. One patient had alveolar hemorrhage and lung infiltration, and eventually died of acute respiratory distress syndrome (ARDS). This patient had initially presented with an infiltrate on chest CT scan with dyspnea and fever. Blood cultures were positive for Candida; initial bronchoalveolar lavage was negative, but a follow-up was positive for Candida. A diagnosis of candidal sepsis and pneumonia was made. Another patient died of progressive metastatic disease and ARDS. This was not considered to be related to everolimus although ARDS per se cannot be ruled out as being related to the drug. No grade 4 cases were seen.
Grade 1 | Grade 2 | Grade 3 | Grade 4 | All Grades | |
|---|---|---|---|---|---|
| Clinical suspicion of pneumonitis, n (%) | 9 (3.3) | 18 (6.6) | 10 (3.6) | 0 | 37 (13.5) |
| Management, n | |||||
| Steroid administration | 0 | 10 | 6 | 0 | 16 |
| Dose adjustment | 2 | 12 | 6 | 0 | 20 |
| Discontinuation of everolimus | 0 | 3 | 7 | 0 | 10 |
| Reversibility | 3 | 11* | 6† | 0 | 20‡ |
Treatment of the 37 patients with clinical suspicion of pneumonitis was as follows (see Table 4). Dose adjustments of everolimus were made for 20 patients, and everolimus was permanently discontinued for 10 patients (including 1 patient with previous dose adjustments). Corticosteroid therapy was initiated in 16 patients, including 14 patients with either dose reductions or discontinuation of everolimus.
Complete resolution was evident for 11 of 18 patients with grade 2 events (with partial reversibility in an additional 2 patients) and for 6 of 10 patients with grade 3 events (with partial reversibility in 1 additional patient). Representative CT scans showing presentation at baseline and subsequent time points for two patients for whom the pulmonary event resolved are shown in Figure 1.
Figure 1. Computed tomography (CT) scan series of representative cases showing resolution of noninfectious pneumonitis. (A) A 74-year-old man with metastatic renal cancer developed a cough approximately 4 months after starting everolimus. Top left: Chest radiograph showed a slight increase in markings in addition to metastases. Top right: Subsequent chest CT scan showed stable metastatic nodules but increased interstitial markings with septal thickening. Bottom left: Improving infiltrates after holding everolimus and a 2-week course of steroids. Bottom right: Resolved infiltrates being maintained on a reduced dose of everolimus. (B) Approximately 3 months after starting everolimus, a 63-year-old man with metastatic renal cancer developed cough and dyspnea with a need for supplemental oxygen. Left: Bilateral ground-glass opacities with mild reticular interstitial disease, predominantly in the lower lobes. Right: After discontinuation of everolimus and high-dose corticosteroids, symptoms and infiltrates cleared.
[More] [Minimize]Regarding the 10 patients with grade 3 events, 3 patients continued treatment with everolimus. One had a dose reduction and all received corticosteroids. Resolution was evident in 1 patient after 2 days, a second patient showed signs of improvement (at the time of data cutoff), and the third died of respiratory failure believed to be due to progression of disease. Of the seven patients who discontinued treatment, complete resolution/reversibility was documented in five cases, and the event was ongoing in one patient at the time of data cutoff. The final patient died with candidal sepsis and pneumonia. (Details of the presentation and management of these patients are given in the online supplement.)
Diagnosis was made clinically in most cases. Bronchoscopy was performed in only 6 of 274 patients (2.2%) receiving everolimus therapy, with lung biopsies performed in 4 (1.5%) patients. Transbronchial biopsy in one patient with grade 3 pneumonitis showed severe acute lung injury with features suggestive of organizing diffuse alveolar damage and rare giant cells. The second patient had grade 2 pneumonitis, and biopsy showed acute fibrinous and organizing pneumonia with rare, poorly formed granuloma. Other detailed reports are not available.
Blinded central radiological review was performed on all available chest CT scans to determine whether new radiographic findings occurred during treatment in patients even without a diagnosis of clinical pneumonitis. Changes consistent with pneumonitis were evident in approximately 20% of baseline CT scans for both treatment groups. Postbaseline review was available for 245 (89.4%) and 132 (96.4%) of patients receiving everolimus and placebo, respectively. New or worsening radiographic changes of pneumonitis were subsequently observed in 53.9% (n = 132) and 15.2% (n = 20) of everolimus- and placebo-treated patients, respectively, for whom scans were available (Table 5). In those without a diagnosis of pneumonitis, 38.9% of patients taking everolimus developed new unexplained radiographic abnormalities compared with 15.2% of patients taking placebo. Symptoms of cough and dyspnea were compared in patients with the presence of new imaging findings confirmed on chest CT scans versus those without new findings (Table 4). Dyspnea was reported in 29.0 versus 25.0%, cough in 20.6 versus 16.2%, and cough and dyspnea in 4.7 versus 10.9%, respectively.
Everolimus 10 mg (n = 274) | Placebo (n = 137) | |
|---|---|---|
| Radiographic studies available for review, n (%) | 245 (89.4) | 132 (96.4) |
| Radiographic findings consistent with noninfectious pneumonitis, n (%) | ||
| At baseline | 47 (17.2) | 26 (19.0) |
| Postbaseline | 144 (52.6) | 36 (26.3) |
| Newly occurring or worsened change, n (%) | 132 (53.9) | 20 (15.2) |
| No clinical pneumonitis | 95 (38.9) | 20 (15.2) |
Results from an exploratory analysis assessing whether the presence of new radiographic findings were correlated with improved efficacy (as determined by progression-free survival, the primary end point of the RECORD-1 trial) indicated that no relationship was present (Figure 2).
Figure 2. Kaplan-Meier probability curve of progression-free survival based on the absence/presence of new radiographic findings consistent with pneumonitis.
[More] [Minimize]Baseline radiographic changes were present in 46 patients (17.2% of all patients taking everolimus) and 9 of 37 patients (24.3%) identified as having clinical pneumonitis. In patients with grade 3 toxicity, however, baseline radiographic changes were found in 5 of 10 patients (50%).
Results of baseline PFTs in patients with and without clinical pneumonitis are shown in Table 6, with a separate analysis of patients with grade 3 pneumonitis. The development of pneumonitis was not found to be associated with more impaired baseline PFTs. In the group without clinical pneumonitis, baseline PFTs in those with and without new radiographic infiltrates on CT showed that the mean percentage (standard deviation [SD]) of predicted forced vital capacity was 83.6 (16.8) versus 83.3 (19.9), respectively; the mean percentage (SD) of predicted diffusing capacity for carbon monoxide was 43.9 (24.6) versus 44.0 (25.4), respectively.
Parameter | No Pneumonitis | Clinical Pneumonitis | Grade 3 Pneumonitis |
|---|---|---|---|
| n | 237 | 37 | 10 |
| FVC | |||
| n | 226 | 34 | 9 |
| Mean, L (SD) | 3.5 (1.0) | 3.8 (0.8) | 3.4 (0.9) |
| Mean % predicted (SD) | 83.1 (19.0) | 88.5 (18.6) | 77.0 (20.0) |
| Median % predicted | 83.8 | 88.1 | 75.7 |
| FEV1 | |||
| n | 227 | 34 | 9 |
| Mean, L (SD) | 2.6 (0.8) | 2.7 (0.6) | 2.6 (0.6) |
| Mean % predicted (SD) | 79.0 (20.1) | 82.7 (18.2) | 74.1 (19.8) |
| Median % predicted | 80.8 | 80.1 | 72.8 |
| DlCO | |||
| n | 175 | 27 | 6 |
| Mean, ml/min/mm Hg (SD) | 13.2 (7.9) | 17.5 (10.0) | 19.3 (6.8) |
| Mean % predicted (SD) | 42.7 (24.7) | 55.7 (26.4) | 59.4 (22.1) |
| Median % predicted | 34.3 | 56.8 | 58.4 |
The mTOR inhibitors are a class of drugs used in the prevention of solid organ transplant rejection and for the treatment of advanced RCC. They are under study for use in other cancers and for treating complications of tuberous sclerosis complex, including lymphangioleiomyomatosis (24). These drugs have been associated with a 5–15% incidence of pneumonitis in solid organ transplant recipients with a wide spectrum of disease severity, varying from subclinical to fulminant (8, 12–14). Dose reduction or interruption has been shown to be helpful in some cases, and corticosteroids have been found to be efficacious in other cases (5, 8, 10, 25, 26). Similar findings have been found with the use of temsirolimus and everolimus in the treatment of cancer, although in this setting, the availability and analysis of serial CT scans during treatment have suggested the occurrence of a higher incidence of asymptomatic radiographic abnormalities (5, 15, 17–19). This may be related to the frequency of CT scans during cancer studies leading to increased detection, use of different dosing regimens, or the routine use of other immunosuppressive agents (including corticosteroids) to prevent rejection in the lung or other solid organ transplant setting.
The presentation of mTOR inhibitor–related pneumonitis is usually with cough and dyspnea, although systemic symptoms of fever and fatigue have been reported in some cases (5, 6, 10, 25, 26). Radiographs have shown areas of ground-glass attenuation and patchy consolidation, and pulmonary function tests have shown either a restrictive pattern or an isolated reduction in diffusing capacity (5, 10, 25, 26).
The etiology of the pneumonitis related to mTOR inhibitors remains speculative. Clinicopathological patterns include interstitial pneumonitis with or without fibrosis, bronchiolitis obliterans organizing pneumonia (8, 10, 12, 27, 28), alveolar hemorrhage (10, 29, 30), and lymphocytic interstitial pneumonia (9). Limited evidence also exists for granulomatous interstitial lung disease, a desquamative interstitial pneumonia or pulmonary alveolar proteinosis pattern, pulmonary vasculitis, and necrosis (29, 31). Helper T-cell lymphocytosis has been reported in bronchoalveolar lavage (10, 32). Although it is not fully understood, hypersensitivity mechanisms are supported by lung biopsies, bronchoalveolar lavage findings, and the observed clinical response to corticosteroids. It has been hypothesized that a cell-mediated autoimmune response may occur if sirolimus exposes cryptic antigens that continue to induce an immune response and subsequent pneumonitis (26). In addition, it is speculated that the high affinity of sirolimus for plasma proteins may render it immunogenic as a hapten with ongoing T-cell recognition of the processed antigen complex (10). Pneumonitis appears be dose dependent in some individuals who tolerate lower doses or in whom the disease abates on reduction in drug dosage. Although there are clinical and pathological similarities of pneumonitis with all mTOR inhibitors, relapse does not always occur after switching to another agent (33).
The rate of investigator-reported pneumonitis in patients with advanced RCC, based on data obtained in the RECORD-1 study in which everolimus therapy was compared with placebo, was 13.5% with everolimus. There were no reports in the placebo group. This rate of clinical pneumonitis is consistent with the literature for temsirolimus, with symptomatic rates of 18% in patients with cancer (5). Drug-induced lung disease can be difficult to establish, as there are no pathognomonic findings. In this study, the diagnosis was made clinically in most cases, with bronchoscopy only occasionally used. This approach is common in patients with advanced cancer and was successful in managing the patients in this study.
Pulmonary toxicity developed from 3.4 to 36.7 weeks (median, 15.4 wk) after starting treatment, similar to other mTOR inhibitors. We found radiographic patterns to be quite variable, from ground-glass infiltrates to more diffuse infiltrates, with the patterns frequently changing over time. As anticipated, chest CT had a higher sensitivity than CXR for the detection of pneumonitis, although the latter improved in sensitivity as the grade of toxicity increased.
Baseline radiographic abnormalities were present in 17% of all of those receiving everolimus and in 24% of those with clinical pneumonitis, but were noted in 50% of patients with grade 3 toxicity. The presence of baseline radiographic abnormalities contributing to an increased risk of serious toxicity previously has been noted for other drugs, notably gefitinib, erlotinib, methotrexate, leflunomide, and temsirolimus as well as with everolimus (19, 34–39). Baseline PFTs did not identify patients with an increased risk of pneumonitis or predict its severity.
Patients identified as having pneumonitis during the study were managed on the basis of suggested guidelines according to severity (Table 1). In our study, 8 of the 10 patients with grade 3 toxicity had a favorable outcome subsequent to treatment according to the management suggestions; of the other 2 patients, 1 died of candidal pneumonia and sepsis, and the other died as a result of progressive RCC and ARDS.
As previously noted with temsirolimus, we found a high frequency of new radiographic findings during everolimus treatment on dedicated CT scan review without a suspected diagnosis of pneumonitis. These were found in 38.9% of patients taking everolimus compared with 15.2% taking placebo. In those patients, no interruption of therapy occurred, and higher-grade toxicity did not develop during the period of observation. The nature and significance of these radiographic findings are not completely known. Transitory radiographic abnormalities are common in those with advanced cancer and multiple comorbidities on surveillance scans. The higher frequency with everolimus may represent a low grade of drug toxicity not clinically detected, but also may reflect other changes occurring in the lungs during cancer therapy, such as capillary leak or an alteration of normal immune responses to inhaled antigens. Knowledge of the occurrence of new radiographic findings that may be transitory during treatment is important clinically to help determine appropriate interventions.
On the basis of our data, we recommend the following essentially noninvasive management for pneumonitis occurring during everolimus treatment, particularly in the context of cancer or RCC:
Grade 1: For patients with radiological changes suggestive of pneumonitis but with few or no symptoms, everolimus therapy may be continued without dose adjustment in most cases. Follow-up with careful clinical assessment should be performed at each visit, with routine radiographic follow-up if asymptomatic. Exceptions should be considered if there is baseline pneumonitis or if the infiltrates are extensive, and these patients could be managed more aggressively. Patients should be specifically cautioned to report respiratory symptoms immediately. Those with baseline radiographic findings should have close interval assessment with chest CT and/or be managed as having grade 2 toxicity.
Grade 2: Patients with moderate symptoms may be managed by temporary interruption or a decrease in dose to 5 mg daily with close observation for resolution of symptoms. Corticosteroids may be helpful for troublesome symptoms if dose alteration is not effective. Diagnostic or therapeutic measures to exclude an infectious origin or other causes of radiographic infiltrates/respiratory symptoms, such as fluid overload or pulmonary embolus, should be performed, preferably before starting corticosteroids. With improvement to an asymptomatic status, everolimus may be reintroduced, if interrupted, at a reduced dose of 5 mg daily with close observation.
Grade 3: For patients whose symptoms are severe, everolimus should be discontinued and the use of corticosteroids may be indicated until clinical symptoms resolve. Therapy with high-dose methylprednisolone in patients with respiratory distress is recommended; lower doses of corticosteroids may be used in less severe cases. It is essential that an appropriate work-up, including bronchoscopy and other diagnostic tests as indicated, be done to exclude any infectious etiology, particularly given the risk of opportunistic infection with the mTOR inhibitors. Other causes such as pulmonary emboli or progressive metastatic disease also should be considered. In general, for any patient with significant respiratory compromise, corticosteroids should be given urgently while the evaluation is proceeding. With resolution of evidence of toxicity, everolimus should be reinitiated at a reduced dose of 5 mg daily in select clinical circumstances with proven advantage and close follow-up.
Grade 4: For patients with life-threatening complications, everolimus should be permanently discontinued. Corticosteroids should be given and an appropriate evaluation to exclude other possible etiologies performed. Everolimus should not be reinstituted even with resolution of pneumonitis.
Both CXRs and CTs are used for monitoring during cancer treatment. The CXR was less sensitive in detecting asymptomatic radiographic findings and also clinical pneumonitis. Sensitivity for detection of pneumonitis did improve with increasing grade of toxicity. CTs have the benefit of earlier detection of all grades of pneumonia and are often helpful in determining the extent of abnormalities and the differential diagnosis. If CXR rather than CT is used for monitoring during treatment, it is essential that a CT be performed for any patient with new respiratory symptoms and a negative CXR and for evaluation of more severe cases of toxicity. A cost–benefit analysis is not available.
Some issues were not addressed in this study. One is the possibility of reescalation of dose with resolution of toxicity on a reduced dose with or without the use of chronic corticosteroids. The role of PFTs in managing toxicity was also not addressed; however, the authors suggest they may be useful in some patients to determine the level of physiological impairment. This would include patients with baseline radiographic changes with the potential for higher level toxicity to help determine the level of impairment in those with extensive radiographic findings and few or no symptoms, and to monitor the effect of interventions such as dose reduction/discontinuation and corticosteroids, in conjunction with imaging. Specific management of patients with baseline radiographic findings was not addressed, but it may be prudent to use closer interval radiographic assessment for this group.
Everolimus has been shown to prolong progression-free survival in patients with advanced RCC (2). Drug-related toxicity can occur during treatment with everolimus and should be considered in patients presenting with nonspecific respiratory signs and symptoms (including hypoxia, cough, or dyspnea) or unexplained radiographic findings and in whom infectious, neoplastic, and other noniatrogenic, causes have been excluded by appropriate investigation. A relatively high incidence of low-grade radiographic abnormalities may be anticipated during treatment. It is important to recognize possible pneumonitis to avoid considering this pulmonary event as progression of disease or infection, and because early recognition and prompt intervention are important in managing the risk associated with pneumonitis. Guidelines are given for this cancer setting; however, this general approach may also be useful for pneumonitis occurring with mTOR inhibitors in other settings.
The following investigators recruited patients to RECORD-1 (listed in alphabetical order by country): Australia—I. Davis, D. Goldstein, H. Gurney, P. Mainwaring, K. Pittman; Canada—S. Ades, T. Cheng, S. Hotte, J. Knox, Y.-J. Ko, M. MacKenzie, S. North; France—A. Caty, C. Chevreau, B. Duclos, B. Escudier, S. Negrier, S. Oudard, A. Ravaud, F. Rolland; Germany—P. Albers, J. Beck, L. Bergmann, V. Grünwald, J. Gschwend; Italy—E. Bajetta, F. Boccardo, S. Bracarda, G. Carteni, P.F. Conte, R. Passalacqua, C. Porta, C. Sternberg; Japan—H. Akaza, M. Eto, H. Fujimoto, Y. Hamamoto, H. Kanayama, M. Maruoka, M. Niwakawa, N. Shinohara, Y. Sumiyoshi, A. Terai, N. Tsuchiya, T. Tsukamoto, H. Uemura, M. Usami; The Netherlands—G. Groenewegen, S. Osanto, F. Van Den Eertwegh, C. Van Herpen; Poland—J. Pikiel, A. Pluzanska, C. Szczylik, R. Zdrojowy; Spain—E. Calvo, D. Castellano, M. Climent, F. del Muro, P. Maroto; United States—C. Alemany, T. Anderson, L. Appleman, J. Beck, W. Berry, M. Danso, A. Dudek, R. Figlin, N. Gabrail, D. George, R. Gersh, M. Gordon, G. Guzley, J. Hajdenberg, J. Hamm, A. Hussain, T. Hutson, P. Lara, D. Loesch, T. Logan, R. Motzer, K. Rathmell, D. Schlossman, D. Smith, J. Thompson, U. Vaishampayan, N. Vogelzang. The authors thank the patients and their families for their participation in the study and Peter Berry (Novartis Oncology) and Victoria A. Robb, Ph.D. (Scientific Connexions), for assistance in the preparation of this manuscript.
| 1. | Taylor AL, Watson CJ, Bradley JA. Immunosuppressive agents in solid organ transplantation: mechanisms of action and therapeutic efficacy. Crit Rev Oncol Hematol 2005;56:23–46. |
| 2. | Motzer RJ, Escudier B, Oudard S, Hutson TE, Porta C, Bracarda S, Grünwald V, Thompson JA, Figlin RA, Hollaender N, et al.; RECORD-1 Study Group. Efficacy of everolimus in advanced renal cell carcinoma: a double-blind, randomised, placebo-controlled phase III trial. Lancet 2008;372:449–456. |
| 3. | Kay A, Motzer R, Figlin R, Escudier B, Oudard S, Porta C, Hutson T, Bracarda S, Hollaender N, Urbanowitz G, et al. Updated data from a phase III randomized trial of everolimus (RAD001) versus PBO in metastatic renal cell carcinoma (mRCC) [abstract 278]. Presented at the Genitourinary Cancers Symposium, February 26–28, 2009, Orlando, Florida. |
| 4. | Atkins MB, Hidalgo M, Stadler WM, Logan TF, Dutcher JP, Hudes GR, Park Y, Liou S-H, Marshall B, Boni JP, et al. Randomized phase II study of multiple dose levels of CCI-779, a novel mammalian target of rapamycin kinase inhibitor, in patients with advanced refractory renal cell carcinoma. J Clin Oncol 2004;22:909–918. |
| 5. | Duran I, Siu LL, Oza AM, Chung T-B, Sturgeon J, Townsley CA, Pond GR, Seymour L, Niroumand M. Characterisation of the lung toxicity of the cell cycle inhibitor temsirolimus. Eur J Cancer 2006;42:1875–1880. |
| 6. | Alexandru S, Ortiz A, Baldovi S, Milicua JM, Ruız-Escribano E, Egido J, Plaza JJ. Severe everolimus-associated pneumonitis in a renal transplant recipient. Nephrol Dial Transplant 2008;23:3353–3355. |
| 7. | Expósito V, Vázquez de Prada JA, Gómez-Román JJ, González-Vilchez F, Llano-Cardenal M, García-Camarero T, Fernández-Valls M, Ruano J, Martín-Durán R. Everolimus-related pulmonary toxicity in heart transplant recipients. J Heart Lung Transplant 2008;27:797–800. |
| 8. | Champion L, Stern M, Israël-Biet D, Mamzer-Bruneel M-F, Peraldi M-N, Kreis H, Porcher R, Morelon E. Sirolimus-associated pneumonitis: 24 cases in renal transplant recipients. Ann Intern Med 2006;144:505–509. |
| 9. | Kahan BD, Knight R, Schoenberg L, Pobielski J, Kerman RH, Mahalati K, Yakupoglu Y, Aki FT, Katz S, Van Buren CT. Ten years of sirolimus therapy for human renal transplantation: the University of Texas at Houston experience. Transplant Proc 2003;35:25S–34S. |
| 10. | Pham P-TT, Pham P-CT, Danovitch GM, Ross DJ, Gritsch HA, Kendrick EA, Singer J, Shah T, Wilkinson AH. Sirolimus-associated pulmonary toxicity. Transplantation 2004;77:1215–1220. |
| 11. | Pérez MJ, Martín RO, García DM, Lozano Rey JM, de la Cruz Lombardo J, Rodrigo López JM. Interstitial pneumonitis associated with sirolimus in liver transplantation: a case report. Transplant Proc 2007;39:3498–3499. |
| 12. | Weiner SM, Sellin L, Vonend O, Schenker P, Buchner NJ, Flecken M, Viebahn R, Rump LC. Pneumonitis associated with sirolimus: clinical characteristics, risk factors and outcome—a single-centre experience and review of the literature. Nephrol Dial Transplant 2007;22:3631–3637. |
| 13. | Morath C, Schwenger V, Ksoll-Rudek D, Sommerer C, Beimler J, Schmidt J, Zeier M. Four cases of sirolimus-associated interstitial pneumonitis: identification of risk factors. Transplant Proc 2007;39:99–102. |
| 14. | Fairbanks KD, Eustace JA, Fine D, Thuluvath PJ. Renal function improves in liver transplant recipients when switched from a calcineurin inhibitor to sirolimus. Liver Transpl 2003;9:1079–1085. |
| 15. | Bellmunt J, Szczylik C, Feingold J, Strahs A, Berkenblit A. Temsirolimus safety profile and management of toxic effects in patients with advanced renal cell carcinoma and poor prognostic features. Ann Oncol 2008;19:1387–1392. |
| 16. | Chan S, Scheulen ME, Johnston S, Mross K, Cardoso F, Dittrich C, Eiermann W, Hess D, Morant R, Semiglazov V, et al. Phase II study of temsirolimus (CCI-779), a novel inhibitor of mTOR, in heavily pretreated patients with locally advanced or metastatic breast cancer. J Clin Oncol 2005;23:5314–5322. |
| 17. | Ellard SL, Clemons M, Gelmon KA, Norris B, Kennecke H, Chia S, Pritchard K, Eisen A, Vandenberg T, Taylor M, et al. Randomized phase II study comparing two schedules of everolimus in patients with recurrent/metastatic breast cancer: NCIC Clinical Trials Group IND.163. J Clin Oncol 2009;27:4536–4541. |
| 18. | Amato RJ, Jac J, Giessinger S, Saxena S, Willis JP. A phase 2 study with a daily regimen of the oral mTOR inhibitor RAD001 (everolimus) in patients with metastatic clear cell renal cell cancer. Cancer 2009;115:2438–2446. |
| 19. | White DA, Schwartz LH, Dimitrijevic S, Di Scala L, Hayes W, Gross SH. Characterization of pneumonitis in patients with advanced non–small cell lung cancer treated with everolimus (RAD001). J Thorac Oncol 2009;4:1357–1363. |
| 20. | Cancer Therapy Evaluation Program, Division of Cancer Treatment and Diagnosis, National Cancer Institute, National Institutes of Health. Common terminology criteria for adverse events [Internet]. Version 3.0; accessed June 1, 2010. Available from: http://ctep.cancer.gov/forms/CTCAEv3.pdf. |
| 21. | Endo M, Johkoh T, Kimura K, Yamamoto N. Imaging of gefitinib-related interstitial lung disease: multi-institutional analysis by the West Japan Thoracic Oncology Group. Lung Cancer 2006;52:135–140. |
| 22. | Crapo RO, Morris AH. Standardized single breath normal values for carbon monoxide diffusing capacity. Am Rev Respir Dis 1981;123:185–189. |
| 23. | Crapo RO, Morris AH, Gardner RM. Reference spirometric values using techniques and equipment that meet ATS recommendations. Am Rev Respir Dis 1981;123:659–664. |
| 24. | Bissler JJ, McCormack FX, Young LR, Elwing JM, Chuck G, Leonard JM, Schmithorst VJ, Laor T, Brody AS, Bean J, et al. Sirolimus for angiomyolipoma in tuberous sclerosis complex or lymphangioleiomyomatosis. N Engl J Med 2008;358:140–151. |
| 25. | Chhajed PN, Dickenmann M, Bubendorf L, Mayr M, Steiger J, Tamm M. Patterns of pulmonary complications associated with sirolimus. Respiration 2006;73:367–374. |
| 26. | Morelon E, Stern M, Israel-Biet D, Correas JM, Danel C, Mamzer-Bruneel M-F, Peraldi M-N, Kreis H. Characteristics of sirolimus-associated interstitial pneumonitis in renal transplant patients. Transplantation 2001;72:787–790. |
| 27. | Carreño CA, Gadea M. Case report of a kidney transplant recipient converted to everolimus due to malignancy: resolution of bronchiolitis obliterans organizing pneumonia without everolimus discontinuation. Transplant Proc 2007;39:594–595. |
| 28. | West ML. Bronchiolitis obliterans and organizing pneumonia in renal transplant recipients. Transplantation 2000;69:1531. |
| 29. | Khalife WI, Kogoj P, Kar B. Sirolimus-induced alveolar hemorrhage. J Heart Lung Transplant 2007;26:652–657. |
| 30. | Singer SJ, Tieman R, Sullivan EJ. Interstitial pneumonitis associated with sirolimus therapy in renal-transplant recipients. N Engl J Med 2000;343:1815–1816. |
| 31. | Schwarz MI, Fontenot AP. Drug-induced diffuse alveolar hemorrhage syndromes and vasculitis. Clin Chest Med 2004;25:133–140. |
| 32. | Howard L, Gopalan D, Griffiths M, Mahadeva R. Sirolimus-induced pulmonary hypersensitivity associated with a CD4 T-cell infiltrate. Chest 2006;129:1718–1721. |
| 33. | De Simone P, Petruccelli S, Precisi A, Carrai P, Doria R, Menichetti F, Filipponi F. Switch to everolimus for sirolimus-induced pneumonitis in a liver transplant recipient—not all proliferation signal inhibitors are the same: a case report. Transplant Proc 2007;39:3500–3501. |
| 34. | Niho S, Kubota K, Goto K, Yoh K, Ohmatsu H, Kakinuma R, Saijo N, Nishiwaki Y. First-line single agent treatment with gefitinib in patients with advanced non–small-cell lung cancer: a phase II study. J Clin Oncol 2006;24:64–69. |
| 35. | Niho S, Goto K, Yoh K, Kim YH, Ohmatsu H, Kubota K, Saijo N, Nishiwaki Y. Interstitial shadow on chest CT is associated with the onset of interstitial lung disease caused by chemotherapeutic drugs. Jpn J Clin Oncol 2006;36:269–273. |
| 36. | Liu V, White DA, Zakowski MF, Travis W, Kris MG, Ginsberg MS, Miller VA, Azzoli CG. Pulmonary toxicity associated with erlotinib. Chest 2007;132:1042–1044. |
| 37. | Saravanan V, Kelly CA. Reducing the risk of methotrexate pneumonitis in rheumatoid arthritis. Rheumatology (Oxford) 2004;43:143–147. |
| 38. | Chikura B, Lane S, Dawson JK. Clinical expression of leflunomide-induced pneumonitis. Rheumatology (Oxford) 2009;48:1065–1068. |
| 39. | Camus P, Kudoh S, Ebina M. Interstitial lung disease associated with drug therapy. Br J Cancer 2004;91:S18–S23. |
