Taxane-induced pulmonary toxicity

INTRODUCTION — The taxanes, paclitaxel (Taxol), nanoparticle albumin-bound paclitaxel (nabpaclitaxel [Abraxane]), docetaxel (Taxotere), and cabazitaxel (Jevtana), are anti-microtubulin drugs that have a broad range of antitumor activity. These agents have the potential to induce pulmonary injury through a variety of mechanisms:

●The most common pulmonary toxicity, interstitial pneumonitis, can develop within days to weeks of receiving paclitaxel or docetaxel, or it may arise later in the course of therapy [1]. In contrast, interstitial pneumonitis seems to be less common with nabpaclitaxel and with cabazitaxel, a semisynthetic taxane derivative used for treatment of advanced prostate cancer.

●A syndrome of capillary leakage, resulting in peripheral edema, noncardiogenic pulmonary edema, and pleural effusions, has also been reported with docetaxel but not with the other taxanes.

●Reactions that occur during or shortly after infusion of an antineoplastic agent can be categorized as hypersensitivity/allergic reactions (associated with mast cell/basophil activation) and standard infusion reactions (cytokine mediated) (table 1). The clinical signs and symptoms often include the lungs, and they overlap, although the mechanisms are different. Patients with symptoms and signs to suggest mast cell/basophil mediation (eg, urticaria, angioedema, wheezing, stridor) are at risk for life-threatening anaphylaxis should rechallenge be undertaken (table 2). The evaluation of reactions during or shortly after infusion of taxanes is presented in detail elsewhere. (See "Infusion reactions to systemic chemotherapy".)

Taxane-induced interstitial pneumonitis and capillary leakage syndrome will be reviewed here. An overview of chemotherapy-induced infusion reactions (which may present with respiratory symptoms) and general issues related to interstitial pneumonitis in adults are presented separately. (See "Infusion reactions to systemic chemotherapy" and "Approach to the adult with interstitial lung disease: Clinical evaluation" and "Approach to the adult with interstitial lung disease: Diagnostic testing" and "Acute interstitial pneumonia (Hamman-Rich syndrome)" and "Idiopathic interstitial pneumonias: Classification and pathology".)

INTERSTITIAL PNEUMONITIS — The interstitial pneumonitis induced by the taxanes, paclitaxel, nanoparticle albumin-bound paclitaxel (nabpaclitaxel), docetaxel, and cabazitaxel, is thought to represent an immune-mediated delayed hypersensitivity reaction [1,2]. In one case report, the patient underwent bronchoalveolar lavage (BAL), and the BAL lymphocytes showed a positive leukocyte migration inhibition test to paclitaxel [3]. Based on this finding, paclitaxel (rather than its vehicle Cremophor) was judged responsible, via a cell-mediated (type IV) delayed reaction (table 3).

Although most cases follow intravenous use of the taxane, fatal acute interstitial pneumonitis has also been described in patients after insertion of paclitaxel-eluting intracoronary stents. (See "Periprocedural complications of percutaneous coronary intervention", section on 'Hypersensitivity reactions'.)

Interstitial lung disease may rarely occur with more chronic dosing, and progression to pulmonary fibrosis has been described [4]. Whether the mechanism underlying chronic interstitial pneumonitis is similar to that responsible for the acute/subacute presentation is unclear.

Paclitaxel, docetaxel, and nabpaclitaxel — There are multiple reports of acute or subacute, diffuse interstitial lung disease developing within hours or a few weeks after paclitaxel [1,5-10] or docetaxel administration [2,11-13]; there are fewer reports detailing pneumonitis after administration of nabpaclitaxel, and most of these events appear to be mild, meaning that they are probably underreported in published clinical trials [14].

Incidence and impact of dose and schedule — Approximately 1 to 5 percent of patients receiving conventional-dose paclitaxel or docetaxel on an every-three-week schedule develop pneumonitis of grade 3 or higher (table 4) severity during therapy [4,13,15-21].

There is a modest dose effect within the range of docetaxel doses that are used in the clinic. This was shown in a phase III trial conducted in females with advanced breast cancer, in which there was a slightly but significantly higher incidence of grade 3 or 4 (table 4) pulmonary toxicity with docetaxel doses of 100 mg/m² as compared with 60 mg/m² (2.2 versus 0.7 percent) [22]. A case of lethal pneumonitis was reported eight days after a patient received a fourth dose of docetaxel at 100 mg/m²; autopsy disclosed severe interstitial fibrosis that was attributed to docetaxel [23].

No differences in rates of pneumonitis have been reported in trials comparing paclitaxel doses within a range that does not require growth factor or hematopoietic stem cell support (eg, 175 versus 225 or 250 mg/m² of paclitaxel [24,25]). However, interstitial pneumonitis has been seen more commonly with higher paclitaxel doses that require growth factor or hematopoietic stem cell support [26,27]. In fact, acute pneumonitis has been dose limiting in phase I trials of paclitaxel [27].

There is some evidence to suggest that weekly as compared with every-three-week dosing of paclitaxel increases the incidence or severity of pneumonitis, although the data are conflicting:

●In a phase III trial directly comparing weekly with every-three-week paclitaxel administration in females with advanced breast cancer, rates of grade 3 or higher (table 4) dyspnea during therapy were 7 and 4 percent, respectively, with weekly or every-three-week treatment. There were two deaths due to pneumonia on the weekly schedule [20].

●Another comparative trial suggests a higher rate of pneumonitis in patients receiving weekly as compared with every-three-week docetaxel [21]. Of the 60 patients receiving weekly therapy, interstitial pneumonia developed in 16 (27 percent), and it was grade 1 or 2 (table 4) in 13 and grade 3 or 4 in three (5 percent). By comparison, interstitial pneumonia only developed in 2 of the 33 patients receiving standard every-three-week therapy (6 percent), only one of which was grade 3.

On the other hand, as noted above, rates of pneumonitis with weekly administration of nabpaclitaxel seem to be lower. A phase III trial compared weekly administration of nabpaclitaxel (100 mg/m²) plus carboplatin every three weeks with non-bound paclitaxel (200 mg/m²) and carboplatin, each given every three weeks in over 1000 patients with non-small cell lung cancer (NSCLC), and reported no significant grade 3 or greater pulmonary toxicity in either arm [28].

Patients with preexisting lung disease appear to have an increased risk of pneumonitis with paclitaxel and docetaxel. In a retrospective series of 392 Japanese patients with NSCLC who were treated with docetaxel (60 mg/m²) every three weeks, the incidence of interstitial lung disease was 4.6 percent, but it was 26 percent in patients with preexisting interstitial lung disease [13]. Two other reports in which patients with underlying interstitial lung disease were treated with docetaxel noted acute exacerbation of the interstitial lung disease in 14 and 18 percent, respectively, with one-half of the reactions resulting in death [29,30].

In our view, preexisting interstitial lung disease should be considered a relative contraindication to taxane treatment. However, at least one case report describes successful treatment with nabpaclitaxel in a patient with NSCLC and a history of interstitial lung disease [31].

Concomitant drugs — The incidence of taxane-induced pneumonitis seems to be higher when taxanes are combined with other cytotoxic agents [1,9,16,32-39]. In particular, the combination of taxanes plus gemcitabine appears to result in significantly more pulmonary toxicity than seen with a taxane alone [9,16,17,32-37,39,40]:

●In one series, 4 of 12 patients (33 percent) with NSCLC who were treated with the combination of paclitaxel and gemcitabine developed grade 2 or worse (table 4) pneumonitis [9]. The diagnosis was made on the basis of high-resolution computed tomography (CT) scanning, and all patients responded to high-dose glucocorticoids within a week or less. No patient had received thoracic radiotherapy (RT) at any time during the course of the illness.

●In two other reports, grade 3 or worse interstitial lung disease was observed in 4 of 39 (10 percent) and 7 of 63 (11 percent) patients treated with the combination of paclitaxel plus gemcitabine or docetaxel plus gemcitabine, respectively [33,41].

●In a meta-analysis, the incidence of severe (grade 3 or worse) lung toxicity with combined docetaxel plus gemcitabine was 2.7 percent (95% CI 2.26-3.14); the proportion of fatal cases was 0.35 percent [40]. Patients treated for lung cancer were at a higher risk for severe pulmonary toxicity than were those treated for breast cancer (4 versus 0.8 percent).

●A report raised the concern that a regimen of docetaxel and ramucirumab after nivolumab therapy may increase the risk of interstitial lung disease in some patients [42].

Concomitant radiotherapy — Taxanes are often used in conjunction with RT and are considered radiosensitizers. Thoracic RT is associated with lung damage, but concomitant therapy with paclitaxel, nabpaclitaxel, or docetaxel and lung irradiation can cause pulmonary disease beyond that attributable to RT alone [43-46]. (See "Radiation-induced lung injury".)

As examples:

●In a trial of concurrent cisplatin, weekly docetaxel, and thoracic RT for NSCLC, pneumonitis of grade 3 or greater severity (table 4) was observed in 4 of 34 (12 percent) patients, while 44 percent had grade 1 or 2 pneumonitis [43].

●A trial of concurrent RT and weekly docetaxel in stage III NSCLC reported a 47 percent rate of grade 3 or worse pneumonitis; the high rate was thought to be related to a large volume of irradiated lung [44].

●A phase II study of nabpaclitaxel plus carboplatin in combination with thoracic radiation reported grade 2 or more severe radiation pneumonitis in all three patients who had more than 30 percent of the lung volume treated to at least 20 Gy [46].

The use of sequential rather than concurrent administration of taxanes and chest wall RT appears to reduce, although not eliminate, the excess risk of pneumonitis [38,47-52]. As an example, a North American trial reported grade 3 to 5 (table 4) pneumonitis in 10 percent of 73 patients (one death) treated with docetaxel consolidation chemotherapy after cisplatin and etoposide plus concurrent chest irradiation [48]. In a separate study, a shorter interval between RT and taxane initiation was associated with a greater risk of pneumonitis [53].

Patterns of disease — Symptom onset varies from a few days after the first dose to weeks following completion of chemotherapy.

Paclitaxel has been associated with acute diffuse interstitial pneumonia (which may progress to acute respiratory failure), subacute diffuse interstitial pneumonitis (with a pattern similar to, but often less severe than, the acute form), pulmonary opacities with peripheral eosinophilia, and pulmonary fibrosis [1,4,54,55]. An organizing pneumonia pattern (bilateral ground glass changes with patchy consolidation in a predominantly subpleural or peribronchial distribution) was a common finding in a small case series [56]. Patients may have respiratory symptoms starting after the first treatment that worsen with subsequent exposures, or may develop symptoms weeks after completion of several cycles [4].

Rare cases of fatal acute interstitial pneumonitis (diffuse alveolar damage) have been described three to seven days after implantation of a paclitaxel-eluting stent [57]. (See "Periprocedural complications of percutaneous coronary intervention", section on 'Hypersensitivity reactions'.)

Docetaxel has been associated with acute or subacute diffuse interstitial pneumonia, usually developing after the second cycle, but occasionally as early as the first cycle or as late as the ninth cycle [13]. This is different from the capillary leakage phenomenon associated with the early onset peripheral edema, noncardiogenic pulmonary edema, and pleural effusions described below. (See 'Capillary leakage and docetaxel' below.)

Clinical manifestations — The clinical manifestations of taxane-induced pneumonitis are nonspecific and include the following:

●Symptoms and physical examination findings – Symptoms of taxane-induced interstitial pneumonitis include dyspnea (at rest or with exertion), dry cough, malaise, and low-grade fever [1,12]. Tachypnea, hypoxemia, and crackles may be present.

●Imaging – The radiographic features of taxane-induced pneumonitis are nonspecific and similar to other cases of drug-induced pneumonitis [1,12,13]. Chest radiographs and high-resolution CT scans typically show an increase in reticular markings in a patchy or diffuse pattern with or without ground glass opacities (image 1) [58-60]. A peribronchial distribution of ground glass opacities has been reported [61]. Focal opacities or dense nodules/consolidation with or without air bronchograms may be seen in patients with pulmonary parenchymal infiltration with eosinophils or organizing pneumonia-type reactions [62]. (See "High resolution computed tomography of the lungs", section on 'HRCT patterns'.)

Diagnostic evaluation — The diagnosis of taxane-induced pneumonitis is usually based on the combination of a compatible clinical pattern, exposure history, and the exclusion of other causes of diffuse pulmonary opacities, such as infection, radiation pneumonitis, alveolar hemorrhage, reaction to a separate antineoplastic agent, pulmonary involvement by the underlying malignancy, heart failure, and noncardiogenic pulmonary edema (for those on docetaxel). (See 'Capillary leakage and docetaxel' below.)

The evaluation will often include laboratory testing, BAL (particularly when systemic glucocorticoid therapy is contemplated), and rarely, lung biopsy. As with other antineoplastic agents, improvement in the interstitial pneumonitis following cessation of taxane therapy supports the diagnosis. The evaluation and diagnosis of interstitial lung disease associated with antineoplastic therapy are discussed separately. (See "Pulmonary toxicity associated with systemic antineoplastic therapy: Clinical presentation, diagnosis, and treatment", section on 'Diagnosis' and "Pulmonary toxicity associated with systemic antineoplastic therapy: Clinical presentation, diagnosis, and treatment", section on 'Differential diagnosis'.)

●Laboratory testing – Laboratory testing is used to determine whether other disease processes are contributory and often includes complete cell counts, coagulation tests, B-type natriuretic peptide (BNP), blood cultures, sputum cultures, and viral and fungal serology and DNA analysis.

●Bronchoalveolar lavage – The main purpose of BAL is to exclude other processes, such as infection, alveolar hemorrhage, or metastatic spread of the underlying cancer. Patients with acute taxane-induced pneumonitis often have lymphocytosis in the BAL specimen; increased eosinophils has also been noted [54,55]. (See "Role of bronchoalveolar lavage in diagnosis of interstitial lung disease".)

●Surgical lung biopsy – Lung biopsy is rarely indicated and usually reserved for patients with progressive lung disease despite cessation of taxane therapy when the cause of the pneumonitis is uncertain.

●Histopathology – Histopathologic findings in taxane-induced pneumonitis include nonspecific interstitial pneumonia, organizing pneumonia, eosinophilic infiltration, and diffuse alveolar damage with intra-alveolar and interstitial foamy macrophages [1,3,12,55,63,64]. (See "Role of lung biopsy in the diagnosis of interstitial lung disease".)

The differential diagnosis of taxane-induced lung toxicity is similar to that of lung toxicity due to other antineoplastic agents and includes infection and lung involvement by the primary malignancy. (See "Pulmonary toxicity associated with systemic antineoplastic therapy: Clinical presentation, diagnosis, and treatment", section on 'Differential diagnosis'.)

For patients on docetaxel, acute pneumonitis needs to be differentiated from acute permeability edema, which is thought to reflect a capillary leakage phenomenon and is associated with peripheral edema. (See 'Capillary leakage and docetaxel' below.)

Treatment — Treatment of taxane-induced pneumonitis is empiric and primarily consists of cessation of taxane therapy. The decision to initiate glucocorticoid therapy usually depends on the severity and rapidity of worsening of pulmonary impairment.

Glucocorticoids are generally reserved for patients with more severe pulmonary toxicity (eg, dyspnea at rest, a decrease in oxygen saturation below 90 percent or a more than 4 percent decrease from baseline, or worsening clinical status). When considering systemic glucocorticoid therapy, it is mandatory to exclude an infectious etiology with appropriate laboratory testing and cultures, often including bronchoscopy and BAL. Empiric antimicrobial therapy directed at likely pathogens is often indicated while diagnostic procedures and cultures are performed.

There is no established glucocorticoid treatment schedule, but severe respiratory compromise is often treated with prednisone 40 to 60 mg daily; intravenous glucocorticoids may be used in patients who have impending respiratory failure [61]. If the patient's response permits, tapering of the dose can be carried out over one to two months.

The rationale for using systemic glucocorticoid therapy in selected patients is based on case reports suggesting an immunologic mechanism (eg, biopsy evidence of hypersensitivity pneumonitis), and anecdotal observation of benefit, particularly in cases that have an acute or subacute presentation [1,12,61,65].

●In patients with an acute or subacute onset of pneumonitis due to paclitaxel or docetaxel, drug cessation and treatment with systemic glucocorticoids usually rapidly reverse the pneumonitis and improve oxygenation [1,6,12]. Three patients with diffuse pneumonitis and hypoxemia (pulse oxygen saturation <90 percent) associated with paclitaxel therapy had complete resolution of symptoms and radiographic opacities with high-dose prednisone [1].

●In a report of two patients with respiratory failure due to taxanes (one after four courses of docetaxel and one after a single course of paclitaxel), both patients experienced resolution of pneumonitis with glucocorticoid treatment [65].

●In a literature review of 31 cases of interstitial pneumonitis associated with docetaxel (either alone or in combination with other drugs, including cyclophosphamide and gemcitabine), pneumonitis improved or resolved with glucocorticoids in 18 and resulted in death in 12 (despite glucocorticoids in 11); outcome was unknown in one [12].

Patients who develop taxane-associated acute interstitial pneumonitis with biopsy evidence of diffuse alveolar damage may progress to life-threatening respiratory failure despite glucocorticoid therapy [2,11,12,66]. Other patients with subacute pneumonitis have developed progressive respiratory failure and death despite drug discontinuation and systemic glucocorticoids [11,67]. (See "Acute interstitial pneumonia (Hamman-Rich syndrome)".)

Cabazitaxel — Interstitial pneumonitis, interstitial lung disease, and acute respiratory distress syndrome have been reported in patients treated with cabazitaxel [68], although the frequency is not known. Patients with underlying lung disease may be at higher risk. The optimal form of treatment is not established. However, the United States prescribing information suggests interrupting therapy for any new or worsening pulmonary symptoms. Corticosteroids may be indicated with severe exacerbations since they have been shown to be of benefit with pulmonary toxicity induced by other taxanes.

CAPILLARY LEAKAGE AND DOCETAXEL — Some patients receiving docetaxel develop serious problems with fluid retention, attributed to capillary leakage. This side effect appears to be most closely related to the total dose of docetaxel administered, but the incidence and severity can be reduced by glucocorticoid pretreatment [69-71].

●Clinical manifestations – Clinically, capillary leakage is usually manifest as peripheral edema and weight gain, although pulmonary vascular leak can also occur, leading to noncardiogenic pulmonary edema and/or pleural effusions [37,72].

●Diagnosis – For patients who develop capillary leakage, the first sign of fluid retention is body weight gain. The diagnosis of noncardiogenic pulmonary edema or pleural effusion is suggested by the time course, concomitant peripheral edema, dyspnea, pulmonary opacities, or pleural effusion on imaging studies, and lack of fever.

●Treatment – Treatment with diuretics at the first sign of weight gain appears to be effective in limiting the severity of docetaxel-induced fluid retention. If the docetaxel is continued and the fluid retention not treated, it may progress to pleural and/or pericardial effusions, pulmonary edema, and ascites.

●Prevention – The optimal premedication schedule has not been determined, but a common practice is to administer dexamethasone premedication (8 mg orally, every 12 hours for six doses, beginning 24 hours prior to docetaxel injection), typically given to prevent docetaxel-related fluid retention [73-75]. This dose is also used to prevent hypersensitivity reactions to docetaxel. Lower dexamethasone doses may be used in patients receiving weekly docetaxel [76]. With premedication, approximately 50 percent of patients experience peripheral edema that can be controlled with diuretics, and the median cumulative dose to onset of moderate to severe fluid retention is approximately 480 mg/m² (range 60 to 780) [77]. (See "Infusion reactions to systemic chemotherapy", section on 'Taxanes'.)

RADIATION RECALL — Rare reports describe radiation recall pneumonitis with paclitaxel and docetaxel. Radiation recall is characterized by the onset of cough and dyspnea associated with radiographic lung opacities that conform to the prior radiation field. (See "Radiation-induced lung injury", section on 'Prior radiation (recall)'.)

●Paclitaxel was associated with recall pneumonitis in a patient who had previously received radiotherapy for a non-small cell lung cancer [78]. A single dose of dexamethasone 20 mg was associated with symptomatic improvement. Premedication was administered with dexamethasone 20 mg orally 24 hours, 12 hours, and immediately before the next dose of paclitaxel, without a recurrence of cough or dyspnea.

●Radiation recall pneumonitis was considered possible in 2 of 19 patients receiving docetaxel and bexarotene chemotherapy [79].

SUMMARY AND RECOMMENDATIONS

●Interstitial pneumonitis

•Diffuse interstitial pneumonitis may present within days (acute) to weeks (subacute) of starting paclitaxel, nanoparticle albumin-bound paclitaxel (nabpaclitaxel), docetaxel, or cabazitaxel, or it may arise later in the course of therapy. (See 'Interstitial pneumonitis' above.)

•The incidence and severity depend on the agent (incidence may be lower with nabpaclitaxel than with paclitaxel), dose, schedule, concurrent chemotherapy agents, and use of concurrent or sequential radiotherapy. Patients with underlying interstitial lung disease have an increased incidence of pneumonitis related to paclitaxel and docetaxel, and we avoid paclitaxel and docetaxel, if at all possible, in these patients. (See 'Paclitaxel, docetaxel, and nabpaclitaxel' above.)

•Clinical features are nonspecific and include exertional dyspnea, dry cough, malaise, and low-grade fever. Tachypnea, hypoxemia, and crackles may be present on examination. Imaging findings are nonspecific. (See 'Clinical manifestations' above.)

•The diagnosis of taxane-induced pneumonitis is based on the combination of a compatible clinical pattern, exposure history, and the exclusion of other causes of diffuse pulmonary opacities. Bronchoalveolar lavage (BAL) is frequently used to exclude infection, hemorrhage, and malignancy. A lung biopsy is rarely needed. (See 'Diagnostic evaluation' above.)

•Treatment of taxane-induced interstitial pneumonitis is empiric and primarily consists of permanent discontinuation of the taxane. Supplemental oxygen is administered as necessary. The decision to initiate glucocorticoid therapy depends on the severity and rapidity of worsening of pulmonary impairment. Before initiating glucocorticoids, it is mandatory to exclude an infectious etiology with appropriate stains and cultures, often including bronchoscopy and BAL. Empiric antimicrobial therapy directed at likely pathogens is often indicated while diagnostic procedures and cultures are performed. (See 'Treatment' above.)

For patients with severe symptomatic pneumonitis (eg, dyspnea at rest, a decrease in oxygen saturation below 90 percent or a more than 4 percent decrease from baseline, or worsening clinical status), we suggest initiating systemic glucocorticoid therapy rather than observation alone (Grade 2B). We generally use oral prednisone 40 to 60 mg daily; intravenous glucocorticoids may be used initially in patients who have impending respiratory failure. If the clinical response permits, tapering of the dose can be carried out over one to two months. (See 'Treatment' above.)

•Many cases will resolve with drug discontinuation and observation, with or without glucocorticoid treatment, although fatalities have been reported. (See 'Treatment' above.)

●Docetaxel-related capillary leakage

•Docetaxel is associated with fluid retention that is dose related and attributed to a capillary leakage phenomenon. The usual initial symptoms are peripheral edema and weight gain, which can progress to noncardiogenic pulmonary edema and/or pleural effusions if untreated. (See 'Capillary leakage and docetaxel' above.)

•We recommend glucocorticoid premedication prior to docetaxel administration to reduce the incidence and severity of fluid retention (Grade 1B). The usual glucocorticoid is dexamethasone, given 8 mg orally, every 12 hours for six doses, beginning 24 hours prior to docetaxel infusion; lower dexamethasone doses may be used in patients receiving weekly docetaxel. (See 'Capillary leakage and docetaxel' above.)

•For patients who develop peripheral edema while receiving docetaxel, we suggest early treatment with diuretics to limit the severity of fluid retention (Grade 2C). (See 'Capillary leakage and docetaxel' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges James Jett, MD, who contributed to an earlier version of this topic review.