INTRODUCTION — Bleomycin is an antitumor antibiotic that was isolated from a strain of Streptomyces verticillus in 1966 [1]. It has been used successfully to treat a variety of malignancies, predominantly germ cell tumors and Hodgkin lymphoma. The major limitation of bleomycin therapy is the potential for life-threatening interstitial pulmonary fibrosis (also called fibrosing alveolitis) in up to 10 percent of patients receiving the drug [2-5]. Other, less common forms of lung injury include organizing pneumonia and hypersensitivity pneumonitis [3].
The pathogenesis and clinical aspects of bleomycin-induced lung injury will be reviewed here. Potential drug interactions that may modify the course of bleomycin-induced lung injury and the therapeutic options available for management will also be discussed. A general approach to immunocompromised patients with respiratory symptoms and the evaluation of interstitial lung disease in patients receiving chemotherapy are presented separately. (See "Approach to the immunocompromised patient with fever and pulmonary infiltrates" and "Pulmonary toxicity associated with systemic antineoplastic therapy: Clinical presentation, diagnosis, and treatment".)
PHARMACOLOGY AND PATHOGENESIS OF LUNG INJURY — The antineoplastic effect of bleomycin is unique among anticancer agents, and is thought to involve the production of single- and double-strand breaks in DNA (scission) by a complex of bleomycin, ferrous ions, and molecular oxygen [2,6,7]. Bleomycin binds to DNA by intercalation of the bithiazole moiety between base pairs of DNA and by electrostatic interactions of the terminal amines. The reduction of molecular oxygen by ferrous ions chelated by bleomycin leads to hydrogen subtraction from the C3 and C4 carbons of deoxyribose, resulting in cleavage of the C3-C4 bond and liberation of a base with a DNA strand break [6]. Bleomycin is inactivated in vivo by the enzyme bleomycin hydrolase, a cytosolic aminopeptidase that has lower activity in the skin and lungs.
The mechanism of bleomycin-induced lung injury is not entirely clear, but likely includes components of oxidative damage, relative deficiency of the deactivating enzyme bleomycin hydrolase, genetic susceptibility, and elaboration of inflammatory cytokines.
Oxidative damage to the lung appears important in the pathophysiology of lung injury, and antioxidants may ameliorate the process [4,8]. Depletion of iron with chelators also reduces the toxicity of bleomycin both in vitro and in vivo, probably secondary to decreased production of free radicals [4,9,10].
Bleomycin hydrolase, an enzyme that degrades bleomycin, is active in all tissues with the exception of the skin and the lung, which may account for the specific toxicity of the drug to these organs [2,6,11]. Studies investigating the varying susceptibility of mouse strains to bleomycin reveal that a bleomycin-resistant strain (BALB/c) has much greater bleomycin hydrolase activity compared to a bleomycin-sensitive strain. This effect is drug-specific, since BALB/c mice are more sensitive to cyclophosphamide-induced lung injury than are some bleomycin-sensitive strains [12].The acute pulmonary injury seen in bleomycin-sensitive mice has been attributed to DNA strand scission with resulting chromosomal injury [13,14].
The chronic fibrotic response to bleomycin-induced injury has been associated with an acquired loss of bleomycin hydrolase activity [15] and is mediated by an immunologic mechanism, characterized by the migration of activated effector cells in the lung and the release of proinflammatory mediators that eventually result in the development of pulmonary fibrosis [16-19]. Nude (athymic) mice are resistant to bleomycin-induced lung injury, suggesting that the inflammatory process is important to the pathogenesis of the disease [20]. In contrast, SCID mice are not resistant, suggesting that T lymphocytes do not play a profibrotic role in the murine model of bleomycin-induced fibrosis [21].
The influence of genetic factors in the development of bleomycin-induced lung injury is suggested by the large variation in bleomycin sensitivity among treated patients, and by data showing that the development of bleomycin-induced lung injury in rodents is strain-specific. Murine strains are recognized as bleomycin-sensitive (C57BL/6) or bleomycin-resistant (BALB/c) according to their propensity to develop bleomycin-induced pulmonary fibrosis [13,14,22].
The importance of inflammatory cytokines in the development of bleomycin-induced lung injury is suggested by several observations:
●Activation of messenger RNA-encoding cytokines after bleomycin exposure in whole lung and bronchoalveolar lavage preparations precedes the proliferative response and the induction of collagen synthesis [16,23,24].
●In animal models, neutralization of the biologic activity of proinflammatory cytokines, with neutralizing antibodies (eg, anti-tumor necrosis factor alpha [TNFa] and anti-transforming growth factor [TGF]-beta), soluble receptors (eg, recombinant human TNFa receptors) or receptor antagonists (eg, interleukin [IL]-1 receptor antagonists) that bind and inactivate corresponding peptides, results in the amelioration of the lung fibrotic process induced by bleomycin [17,25,26]. Furthermore, animals in which the TNFa receptors have been deleted by recombinant technology are protected from the development of bleomycin-induced lung injury and fibrosis [18].
●Mice that lack the gene encoding cytosolic phospholipase A2, a key enzyme in the generation of proinflammatory thromboxanes and leukotrienes, have an attenuated fibrotic response following bleomycin administration [27].
The alveolar macrophage (AM) is thought to play a central role in the development of bleomycin-induced lung injury due to its ability to release a number of effector molecules (eg, cytokines, lipid metabolites, oxygen radicals). The mechanism for activation of the AM by bleomycin is unknown. Bleomycin receptors have been identified on the surface of rat AMs, suggesting that activation might be mediated through a second messenger [28].
Another report, using mice engrafted with bone marrow isolated from transgenic mice expressing enhanced green fluorescent protein, found that bleomycin treatment induced migration of fibroblasts from the bone marrow to the lung [29]. These fibroblasts express type I collagen, telomerase, and CXCR4 and CCR7 chemokine receptors.
INCIDENCE — The potential for bleomycin-related pulmonary toxicity has been appreciated for many decades, but the scope of the problem is not well documented. The available data on incidence come predominantly from patients treated with a bleomycin-containing regimen for a testicular or ovarian germ cell tumor or ovarian sex cord-stromal tumor, or Hodgkin lymphoma.
Germ cell and ovarian sex cord stromal tumors — In trials of standard-dose chemotherapy for testicular or ovarian germ cell tumors and for ovarian sex cord-stromal tumors (three or four cycles of BEP [bleomycin, etoposide, cisplatin], PVB [cisplatin, vinblastine, bleomycin] or CVB [carboplatin, vinblastine and bleomycin]), which contain a cumulative bleomycin dose of 270 International Units (IU; 135 IU/m2, three courses) or 360 IU (180 IU/m2, four courses), rates of any grade of pulmonary toxicity range from 5 to 16 percent, and rates of fatal pulmonary toxicity have been in the range of 0 to 1 (for three courses) and 0 to 3 percent (for four courses), respectively [30-42]. In at least one report, long-term pulmonary toxicity (unspecified) persisted in 8 percent of patients treated with three courses of BEP [31]. With few exceptions [31,32,36], most of these reports defined bleomycin pulmonary toxicity according to symptoms and clinical findings of pneumonitis and respiratory failure, and not by systematic assessment for asymptomatic pulmonary toxicity using pulmonary function tests (PFTs). (See 'Screening asymptomatic patients for lung toxicity' below.)
However, more recent data suggest that the overall risk of clinically-apparent and fatal bleomycin-induced lung injury may be lower, and that pulmonary toxicity is reversible and without long-term sequelae in most cases. The best data come from a registry-based series of 565 consecutive patients derived from the Danish Testicular Cancer database, who were treated with BEP for a germ cell tumor and underwent systematic and close monitoring of PFTs before, during, and for five years after treatment [43]. The maximum cumulative dose of bleomycin was limited for most patients to 150 IU/m2, which is not currently standard practice for patients with intermediate- and poor-risk germ cell tumors. (See "Initial risk-stratified treatment for advanced testicular germ cell tumors", section on 'Intermediate- and poor-risk advanced disease'.)
Overall, more than 50 percent of the patients had a history of smoking. Approximately 9 percent discontinued bleomycin because of a significant (≥25 percent) decline in hemoglobin-corrected diffusing capacity of the lungs for carbon monoxide (DLCO) during treatment. Other pulmonary function tests (forced expiratory volume in one second [FEV1] and forced vital capacity [FVC]) were unchanged after BEP therapy. Excluding pulmonary embolism, 12 patients (2.1 percent) developed nonfatal acute pulmonary toxicity that was potentially attributable to bleomycin, and only one of the 15 deaths during treatment was attributed to bleomycin pneumonitis (0.17 percent).
At five years of follow-up of the Danish Testicular Cancer database, only a minority of BEP-treated patients, including those who discontinued bleomycin due to a decrease in DLCO, suffered long-term restrictive disease (4.1%; 95% CI 1.8-6.3%) or obstructive lung disease (2.7%; 95% CI 0.8-4.6%) [43]. After a median follow-up of 16.1 years, the 15-year cumulative risk for pulmonary disease in treated patients was low, and comparable to that of a contemporaneously treated group of 2548 patients with stage I germ cell tumors who received no chemotherapy and were observed in a surveillance program (table 1). Although pulmonary function returned to normal in the majority of patients over time, patients who underwent pulmonary surgery or suffered pulmonary embolism during or before treatment had continuously decreased pulmonary function as did smokers. Importantly, there were no differences in long-term outcome among patients who received all of the doses of bleomycin (median 142 IU/m2), versus those whose doses were attenuated because of changes in the DLCO (median dose 100 IU/m2).
These data suggest that identification of patients who have asymptomatic decreases in DLCO, and early discontinuation of bleomycin in such individuals leads to low rates of long-term bleomycin-induced pulmonary disease without compromising oncologic outcomes. (See 'Screening asymptomatic patients for lung toxicity' below.)
Patients with primary mediastinal nonseminomatous germ cell tumors (PMNSGCTs) who require postchemotherapy thoracic surgery to remove residual disease may be more susceptible to postoperative respiratory complications related to bleomycin, however the data are inconsistent [44-46]:
●In one report, among 221 patients with PMNSGCT who underwent postchemotherapy mediastinal or lung surgery at a single institution, acute respiratory failure or pneumonia developed in 22 of the 166 patients who had received BEP (13 percent) and none of the 55 who had received etoposide, ifosfamide, and cisplatin (VIP) [44].
●On the other hand, in a smaller single-center report, there was no significant difference in pulmonary complications among patients with PMNSGCT treated with or without bleomycin. In addition, no fatal pulmonary complications were observed [46].
Therefore, UpToDate authors remain comfortable treating such patients with BEP in the absence of other risk factors for bleomycin pulmonary toxicity and with serial PFTs to monitor DLCO and vital capacity. To minimize postoperative risks in all patients previously treated with BEP including those with PMNSGCT, limitation of supplemental oxygen and avoidance of excess intravenous fluid support is recommended. (See 'Supplemental oxygen and future perioperative management' below and "Extragonadal germ cell tumors involving the mediastinum and retroperitoneum", section on 'Mediastinal nonseminomatous GCTs'.)
Hodgkin lymphoma — Depending on the definition used and the cumulative bleomycin dose, rates of bleomycin-induced pulmonary toxicity in adults receiving ABVD (doxorubicin, bleomycin [20 IU/m2 per course], vinblastine, dacarbazine) for Hodgkin lymphoma range from 10 to 53 percent, and rates of fatal pulmonary toxicity are 4 to 5 percent [47-51]. Long-term functional respiratory impairment is reported in approximately 15 to 18 percent of patients [48,52,53].
Higher rates of toxicity, particularly fatal pulmonary toxicity, have been noted among elderly patients (see 'Age' below). As examples:
●In one report of patients 60 years of age or older who were enrolled in one of two trials comparing two courses of ABVD or AVD, each followed by involved field radiotherapy versus four cycles of ABVD followed by involved field RT, bleomycin lung toxicity was uncommon in those whose treatment was limited to two cycles of ABVD (two cases per 137 patients), but it developed in 7/68 patients (10 percent) who received four courses of ABVD, and was fatal in three (4.4 percent) [54].
●In another series of 95 elderly patients with Hodgkin lymphoma, who received a bleomycin-containing regimen (number of cycles not reported), the incidence of bleomycin lung toxicity was 32 percent, and the associated mortality rate was 25 percent [55].
●Similar findings were noted in another series of 147 individuals with Hodgkin lymphoma age 60 and over who were treated with at least one course of ABVD in three French centers [56]. The median number of courses was six (range one to eight). Granulocyte colony-stimulating factors was administered to 81 (55 percent). (See 'Colony stimulating factors' below.)
ABVD was stopped in 25 patients, and modified in 31; among the 31 patients with modified ABVD, bleomycin was dose reduced in six because of pulmonary toxicity, and deleted from the regimen in 22 for the same reason. Of the 63 patients with grade 3 or 4 toxicities during treatment, 31 were pulmonary (21 percent). There was no significant correlation between bleomycin toxicity and underlying lung disease, tobacco history, age, use of G-CSF, or radiotherapy. Of the 15 deaths from acute toxicity, seven were related to pulmonary toxicity (incidence of grade 5 pulmonary toxicity 4.8 percent).
These findings have led to the general recommendation that elderly individuals with Hodgkin lymphoma be closely monitored with pulmonary function testing after two cycles of ABVD and that bleomycin be omitted in any patient who develops symptomatic or asymptomatic pulmonary toxicity during treatment. (See "Initial treatment of advanced (stage III-IV) classic Hodgkin lymphoma", section on 'Older adults'.)
RISK FACTORS — Age, cumulative drug dose, renal function, the severity of the underlying malignancy at presentation, and also concomitant use of oxygen, radiation therapy, other chemotherapeutic agents, and hematopoietic colony stimulating factors may all influence the risk of developing bleomycin lung toxicity (table 2) [3,5,47]. For older adult patients and those with renal insufficiency, if an equally effective alternative is available (such as for treatment of a germ cell tumor), a bleomycin-free regimen is generally preferred. (See "Initial risk-stratified treatment for advanced testicular germ cell tumors" and "Treatment of malignant germ cell tumors of the ovary".)
Age — In most series, the risk of bleomycin-induced lung toxicity appears higher in older patients:
●The United Kingdom Royal Marsden NHS Trust reported that among 835 patients treated with bleomycin-containing regimens for germ cell tumors, the rate of pulmonary complications was more than twice as high among those over 40 years of age (HR = 2.3) [5].
●In a study of 141 patients with Hodgkin lymphoma who received regimens including bleomycin, the mean age of those with and without bleomycin toxicity was 49 versus 29 years [47].
●A Scottish study of 194 patients treated with bleomycin for germ cell tumors reported that the median age of the five patients who died of pulmonary toxicity was 55 compared to a median age of 33 among those who did not develop fatal pulmonary toxicity [57].
Other studies have not reported an association between older age and impaired pulmonary function tests after bleomycin treatment, although the age range in one of these studies was only 26 to 40 [43].
Dose and renal insufficiency — The incidence of bleomycin-induced pulmonary fibrosis is largely dependent on cumulative drug dose [2,30-33,52]. In patients exposed to a total of 270 international units (IU) or less (one IU = 1 mg), high-grade lung toxicity is seen in 0 to 2 percent, while rates among patients receiving doses of 360 units or more range from 6 to 18 percent [2,30-33]. Cumulative doses >400 units are associated with higher rates of pulmonary toxicity, and are generally avoided [58]. Although high-grade lung injury is very rare with cumulative doses under 400 units, injury can occur at doses less than 50 units. Rapid intravenous infusion may also increase the risk of toxicity [59,60].
Renal insufficiency is a risk factor for bleomycin toxicity [5,61]. This is not surprising since more than 80 percent of the drug is eliminated by the kidney in normal individuals [62]. As an example, the United Kingdom Royal Marsden NHS Trust reported that among 835 patients treated with bleomycin-containing regimens for germ cell tumors the rate of pulmonary complications was over three times higher among those with a glomerular filtration rate <80 mL/minute [5].
Other chemotherapy and radiation therapy — In the treatment of germ cell and ovarian sex-cord-stromal tumors, bleomycin is administered in conjunction with other chemotherapy agents, including cisplatin. At least some data suggest that high cumulative doses of cisplatin also contribute to late impairment of pulmonary function and restrictive lung disease in long-term testicular cancer survivors [63], although others have not found such an association [43]. (See "Treatment-related toxicity in men with testicular germ cell tumors", section on 'Late chemotherapy toxicity'.)
Concomitant administration of gemcitabine with bleomycin may also increase the risk of pulmonary toxicity [64].
In many (but not all [52,56]) reports, thoracic irradiation increases the risk of bleomycin lung toxicity, whether it is administered prior to, or simultaneously with bleomycin. It is unclear whether a long interval between radiation therapy (RT) and administration of bleomycin eliminates the increased risk of lung injury. However, preliminary evidence from a study of 15 patients with advanced-stage Hodgkin lymphoma suggests that the risk of pulmonary toxicity during consolidative RT is low when there is an interval of at least four weeks between chemotherapy and initiation of RT [64].
High fraction of inspired oxygen — Data are conflicting about whether exposure to high concentrations of inspired oxygen (FiO2) increases the likelihood of lung toxicity in patients who have received bleomycin chemotherapy. In animal models, simultaneous exposure to bleomycin and high FiO2 increases the risk of bleomycin-induced lung injury, whereas exposure to lower than normal oxygen concentrations appears protective [65-67].
Whether any of these data apply to humans is not clear. The evidence that oxygen exposure may increase the risk of pulmonary toxicity in humans is largely anecdotal:
●Acute respiratory failure from acute respiratory distress syndrome (ARDS) has been reported following general anesthesia in patients previously treated with bleomycin [68-70]. Following the death of five bleomycin-treated patients from postoperative pulmonary complications at a single institution, a new intraoperative protocol was developed in which oxygen exposure was minimized and intravenous fluid replacement was judiciously administered [68]. With adoption of this protocol, none of the subsequent 12 patients who underwent post-bleomycin surgery for metastatic germ cell tumors developed pneumonitis or died from postoperative pulmonary complications.
●Several subsequent reports documented that exposure to supplemental oxygen was followed by a recurrence of lung injury in patients who had previously had bleomycin pneumonitis; although in some cases the inspired fraction of oxygen was as low as 33 percent [70,71].
Not all of the data regarding the effect of inspired oxygen on bleomycin lung injury are consistent, however. In a review of 316 patients who had previous bleomycin chemotherapy and underwent a surgical procedure lasting at least one hour, ARDS developed in seven [72]. No difference was noted in the mean or peak FiO2 between those who developed ARDS and those who did not. Similarly, a review of 77 patients undergoing major surgery following bleomycin-containing chemotherapy failed to demonstrate a correlation between perioperative oxygen restriction and either postoperative pulmonary morbidity or mortality [73]. These authors suggested that careful fluid management during surgery was more important than limiting FiO2.
Nonetheless, the anecdotal data in humans, combined with the animal data, have had a dramatic impact on clinical practice and have led to widespread recommendations for lifelong avoidance of high concentrations of supplemental oxygen in patients previously exposed to bleomycin unless necessary to maintain an adequate arterial oxygen saturation. It is unknown whether a threshold fraction of inspired oxygen or duration of therapy exists above which the risk of lung injury rises, or whether there is an interval following bleomycin treatment after which higher oxygen concentrations will not increase the risk of lung injury. (See "Adverse effects of supplemental oxygen".)
Cigarette smoking — Cigarette smoking is reported to be a risk factor for bleomycin pulmonary toxicity in some [4,43,74,75], but not all studies [48,56,61,76-78]. The discrepancy may be due to other confounding factors such as bleomycin dose, renal function, patient age, use of other chemotherapeutic agents, and thoracic irradiation.
However, cigarette smoking may be a risk factor for development of ARDS following major surgery among patients previously treated with bleomycin. In a review of 316 patients who had received bleomycin chemotherapy and subsequently underwent a surgical procedure, current or former cigarette smoking was associated with an increased risk for postoperative ARDS [72]. (See 'High fraction of inspired oxygen' above.)
Colony stimulating factors — Concomitant treatment with granulocyte colony-stimulating factor (G-CSF; filgrastim) during bleomycin-containing chemotherapy was identified as a possible risk factor for the development of bleomycin-induced lung injury in animal studies. However, data in humans are conflicting [11,47,50,56,79-83]. One reason for the disparate results may be the confounding influence of age. (See 'Incidence' above.)
Regardless, many clinicians avoid using G-CSF during treatment with regimens containing bleomycin, particularly ABVD (doxorubicin, bleomycin, vinblastine and dacarbazine). (See "Initial treatment of advanced (stage III-IV) classic Hodgkin lymphoma".)
In germ cell tumors, rather than avoiding G-CSF use altogether during bleomycin chemotherapy, investigators have instead successfully altered the schedule to avoid G-CSF administration on the same day as bleomycin infusion [35].
Inflammatory markers — Increases in inflammatory cytokines are a prominent component of the pathogenesis of bleomycin pulmonary toxicity. (See 'Pharmacology and pathogenesis of lung injury' above.)
This raises the possibility that laboratory markers of inflammation might predict the development of bleomycin lung injury. In a retrospective study of 57 patients with germ cell tumors, of whom 15 developed bleomycin pneumonitis, a pre-chemotherapy neutrophil-to-lymphocyte ratio ≥6 and a Prognostic Nutritional Index score of <40 were independently associated with bleomycin pulmonary toxicity, even when accounting for age and smoking history [84]. While of interest, these data were derived from a single, small retrospective study without a uniform objective definition of bleomycin lung injury. As such, further evaluation of these markers is needed.
SCREENING ASYMPTOMATIC PATIENTS FOR LUNG TOXICITY — Practices vary in terms of screening for lung toxicity during bleomycin therapy. For patients without risk factors for bleomycin toxicity (table 2), screening may be limited to assessment of symptoms and having a low threshold for stopping bleomycin if dyspnea or nonproductive cough develop. Pulmonary function testing and 18-fluorodeoxyglucose-positron emission tomography (FDG-PET) can identify early toxicity in certain settings, although supportive data are limited.
Pulmonary function tests — We suggest assessment of pulmonary function tests (PFTs), typically spirometry and diffusing capacity for carbon monoxide (DLCO), at baseline prior to treatment and at intervals during therapy for most patients with germ cell tumors and older adults with Hodgkin lymphoma receiving a bleomycin-containing chemotherapy regimen. While this practice may not be necessary for all patients, it should be strongly considered for those with a history of pulmonary disease, baseline computed tomography suggestive of underlying pulmonary disease, or one or more risk factors for bleomycin pulmonary toxicity [85]. (See 'Risk factors' above.)
There are certain nuances to the interpretation of the DLCO (eg, adjustment for anemia and reduced inspiratory capacity) that are discussed below. (See 'Pulmonary function testing' below.)
Pretreatment PFTs serve as a baseline for future comparisons. Some clinical trials have used baseline PFTs to assess eligibility for receipt of bleomycin, excluding patients with a DLCO (adjusted for hemoglobin) of <40 to 60 percent of predicted (depending on the study). The rationale is that patients with poor baseline DLCO may be more likely to develop symptomatic or life-threatening pulmonary toxicity with even a small degree of bleomycin injury, although prospective data demonstrating the validity of this approach are lacking. Additional problems with interpretation of a baseline DLCO include reductions in inspiratory capacity due to airway compression by mediastinal masses and the inability to put forth maximal inspiratory effort due to chest pain. Under these circumstances, we do not generally exclude patients from receiving bleomycin based on a low DLCO in the absence of other evidence of lung parenchymal disease.
The optimal frequency of testing during bleomycin therapy is not established. For patients with germ cell tumors, we suggest monitoring PFTs prior to each new treatment cycle. For older adults with Hodgkin disease receiving a bleomycin-containing regimen such as ABVD (doxorubicin, bleomycin, vinblastine, dacarbazine), we monitor closely for symptoms of lung toxicity and consider repeating PFTs after every two to four cycles of therapy or at the onset of symptoms. We suggest that bleomycin be discontinued if there is a decrease in the DLCO of 25 percent or more, even if asymptomatic. We do not routinely perform screening chest radiographs in asymptomatic individuals.
There is no widespread consensus as to the utility of screening asymptomatic patients for early signs of bleomycin-induced lung disease, and practice is variable, as evidenced by the following:
●At most institutions, patients undergoing bleomycin-therapy for an adult germ cell malignancy or ovarian sex-cord stromal tumor undergo a baseline set of PFTs followed by periodic retesting during treatment. On the other hand, most clinicians do not routinely assess PFTs in asymptomatic adults receiving a bleomycin-containing regimen for Hodgkin disease with the exception of elderly patients. (See "Initial treatment of advanced (stage III-IV) classic Hodgkin lymphoma", section on 'Older adults'.)
●The US Food and Drug Administration (FDA)-approved prescribing information recommends that DLCO be monitored monthly if it is employed to detect toxicities and that the drug should be discontinued when the DLCO falls below 30 to 35 percent of the pretreatment value. By comparison, in the Danish Testicular Cancer database study, bleomycin was discontinued if the DLCO (adjusted for hemoglobin) declined by 25 percent or more below the baseline level [43]. Other studies in germ cell tumors have also used this cutoff [86,87]. The calculation to adjust the DLCO for hemoglobin is described separately. (See 'Pulmonary function testing' below and "Diffusing capacity for carbon monoxide", section on 'Anemia'.)
Although the FDA-approved prescribing information also recommends a chest radiograph every one to two weeks to monitor the onset of pulmonary toxicity, this is not routinely done.
●Guidelines from the National Comprehensive Cancer Network (NCCN) [88] recommend that baseline pulmonary function tests be "considered" and repeated only as clinically indicated. However, the NCCN does not clarify how the results of such baseline testing are to be used.
●British guidelines support baseline PFTs, although they state that the results should not be used to predict those likely to develop toxicity [85]. They also state that PFTs during treatment may aid in the diagnosis of suspected toxicity and may guide management of toxicity.
This variability in practice likely stems from conflicting data on the use of serial PFTs (eg, spirometry and DLCO) to identify incipient lung toxicity during treatment:
●In early clinical trials, decreases in lung volumes and DLCO appeared to precede the development of severe bleomycin lung damage, and a decline in DLCO appeared to be the earliest and most sensitive indicator of subclinical lung injury [89,90].
●Subsequent studies suggested that PFTs, including DLCO, are neither sensitive nor specific for bleomycin lung toxicity, and many questioned the clinical significance of these changes [4,91-94]. At least one randomized trial concluded that changes in PFTs during bleomycin therapy were only weakly correlated with increased toxicity, and only at the end of treatment [95].
●On the other hand, the data reported from the contemporary Danish Testicular Cancer database suggest that a systematic approach to assessing spirometry and DLCO before and after therapy with early discontinuation of bleomycin for those with a drop in the DLCO of at least 25 percent results in very low rates of both acute and chronic lung disease without affecting oncologic outcomes [43]. In this study, 9 percent of patients discontinued bleomycin due to a decrease in DLCO, but subsequently experienced improvement in DLCO into the range of other participants. The mean dose of bleomycin was 142 IU/m2 among those completing therapy and 100 IU/m2 among those in whom bleomycin was stopped early. Persistently reduced lung volumes and DLCO were noted in 4 percent. Bleomycin-induced lung toxicity requiring glucocorticoids was noted in two patients, and late (15 years after therapy) appearance of pulmonary fibrosis was reported in three patients.
Another point of disagreement is the threshold of DLCO decrease below which bleomycin should be discontinued. One reason for this disagreement is the concern that routine DLCO screening for pulmonary toxicity results in many false positive results, which may lead to the premature and unnecessary discontinuation of bleomycin. The clinical consequences include inferior cancer control and/or toxicity due to the substitution of an alternative, potentially more toxic agent for the bleomycin. However, the data from the Danish Testicular Cancer database do not bear out these concerns, suggesting that, cancer control is not inferior, and that pulmonary outcomes are better if bleomycin is discontinued when the DLCO drops by 25 percent or more below the baseline value [43]. (See 'Germ cell and ovarian sex cord stromal tumors' above.)
The following recommendations inform this debate:
●Several authors recommend that bleomycin be discontinued if the DLCO falls during therapy, although the threshold for discontinuation in the published literature varies, with most suggesting a threshold drop in DLCO of 15 to 35 percent warrants treatment discontinuation [40,43,48,96-99].
●The FDA-approved prescribing information recommends discontinuation of drug when the DLCO falls below 30 to 35 percent of the pretreatment value.
●Consensus-based guidelines from the NCCN [88] do not address the DLCO threshold at which bleomycin should be discontinued.
●No guidelines provide a recommendation for any PFT parameter other than DLCO.
For patients who experience a decline in the DLCO of 25 percent or more during treatment, we suggest discontinuation of bleomycin. In addition, an evaluation for another explanation for decreased lung function may be appropriate depending on the response to drug discontinuation. (See 'Permanent discontinuation of bleomycin' below and 'Evaluation of patients with suspected pulmonary toxicity' below.)
Role of FDG-PET — Monitoring of uptake on 18-fluorodeoxyglucose-positron emission tomography (FDG-PET) scans has been studied as a potential screening tool for early detection of bleomycin-induced pulmonary injury, but the data are limited, and this is not yet a standard approach for most patients receiving bleomycin, particularly in the setting of germ cell tumor treatment [51,100-102]. The FDG-PET appearance in bleomycin toxicity is typically low-level, diffuse, bilateral, often subtle uptake in the lower lobes, which can predate abnormalities on high-resolution CT [103]. While the FDG-PET pattern is not sufficiently specific to make a diagnosis, any FDG uptake in the lungs is abnormal and requires further evaluation. (See 'Imaging' below.)
For patients with Hodgkin lymphoma who are receiving a bleomycin-containing combination such as ABVD, interim FDG-PET computed tomography (PET-CT) is often obtained to restage disease activity for the purpose of tailoring subsequent treatment. (See "Initial treatment of advanced (stage III-IV) classic Hodgkin lymphoma", section on 'Response adapted therapy'.)
While the focus of the scan is on disease activity, bleomycin should be discontinued if the interim scan shows increased FDG uptake in the lungs suggestive of bleomycin lung toxicity; this position is supported by others as well [103]. In a retrospective analysis of 77 consecutive patients with Hodgkin lymphoma who had interim or end-of-treatment PET scans, the authors identified possible bleomycin lung toxicity in 13 (17 percent) patients, eight of which were identified at the early interim scan; six were asymptomatic [51]. All patients were treated with methylprednisolone and cessation of bleomycin. Only one, who was asymptomatic at the time of the scan, subsequently died of bleomycin toxicity while the others had resolution of the abnormality and did not progress to pulmonary fibrosis. (See 'Imaging' below.)
CLINICAL PRESENTATION — The range of clinical presentations of bleomycin-induced lung injury includes symptoms or physical examination findings (eg, dyspnea, cough, chest pain, and crackles on physical examination), the presence of opacities on chest radiographs, or an asymptomatic decline in diffusing capacity for carbon monoxide (DLCO) during bleomycin therapy. Uncommonly, bleomycin-induced lung injury may be diagnosed several years after bleomycin exposure as chronic fibrotic lung disease or the acute onset of respiratory impairment in the postoperative period.
Symptoms of bleomycin-induced lung injury usually develop subacutely (over days to weeks) within one to six months of beginning bleomycin treatment, but may rarely occur more than six months following the administration of the drug [104]. Symptoms and physical signs associated with bleomycin-induced lung injury are nonspecific with the earliest symptom being dyspnea and the earliest sign being auscultatory crackles (table 3) [4].
For other types of lung injury, the time course of onset and pace of progression of clinical manifestations may suggest a particular type of lung injury:
●Bleomycin-induced hypersensitivity pneumonitis and diffuse alveolar damage typically present with more rapidly progressive symptoms.
●An acute chest pain syndrome occurs in approximately 1 percent of patients during infusion of bleomycin, but does not predict the development of pulmonary fibrosis [105]. (See "Infusion reactions to systemic chemotherapy", section on 'Bleomycin'.)
●An indolent onset of dyspnea on exertion several months after completion of bleomycin therapy is suggestive of the fibrotic form of bleomycin lung toxicity, similar to usual interstitial pneumonia. (See "Clinical manifestations and diagnosis of idiopathic pulmonary fibrosis", section on 'Clinical manifestations'.)
EVALUATION OF PATIENTS WITH SUSPECTED PULMONARY TOXICITY — Evaluation of patients with clinical signs and/or symptoms (especially cough) of suspected bleomycin-induced lung injury or an asymptomatic drop in the diffusing capacity for carbon monoxide (DLCO) typically includes a complete blood count with differential, radiographic imaging, and pulmonary function testing. Many patients will also undergo bronchoscopy with bronchoalveolar lavage to rule out infection or malignancy. Lung biopsy is typically reserved for patients in whom the diagnosis remains unclear after initial testing.
Laboratory — Occasionally, a mild peripheral blood eosinophilia will be noted in patients with hypersensitivity pneumonitis [106]. For patients with testicular or ovarian germ cell tumors, certain blood tests (eg, human chorionic gonadotropin [hCG] and alpha fetoprotein [AFP], lactate dehydrogenase [LDH]) may be used to assess whether metastasis to the lungs might be an alternate explanation for the findings suggestive of interstitial lung disease, although interstitial pulmonary involvement by germ cell tumors is rare. (See 'Diagnosis and differential diagnosis' below and "Serum tumor markers in testicular germ cell tumors" and "Diagnosis and treatment of relapsed and refractory testicular germ cell tumors", section on 'Diagnosis of relapsed disease' and "Ovarian germ cell tumors: Pathology, epidemiology, clinical manifestations, and diagnosis".)
For patients who are at risk for myocardial dysfunction, a serum brain natriuretic peptide (BNP) level may help identify heart failure. (See "Heart failure: Clinical manifestations and diagnosis in adults".)
Imaging — A chest radiograph is usually obtained to evaluate symptoms, such as dyspnea, cough, or chest pain, or signs such as crackles on lung exam, or hypoxemia. The appearance of bleomycin-induced lung injury on chest radiographs is variable (table 4). The classic pattern of bleomycin-induced pulmonary fibrosis includes bibasilar subpleural reticular opacification with volume loss and blunting of the costophrenic angles; however, fine nodular densities may also be present. These early findings may evolve to progressive consolidation and honeycombing. (See "Evaluation of diffuse lung disease by conventional chest radiography".)
Pneumothorax and/or pneumomediastinum are rare complications of bleomycin-induced pulmonary fibrosis [107].
High-resolution computed tomography (HRCT) of the chest is used to characterize the pattern, location, and extent of abnormalities noted on a chest radiograph and to evaluate gas transfer abnormalities noted on pulmonary function tests. HRCT is generally not used as a screening tool for bleomycin-induced lung injury, although HRCT is more sensitive than chest radiograph in identifying lung abnormalities in bleomycin-exposed patients [108].
HRCT patterns associated with bleomycin toxicity usually reflect the underlying histopathology [109]. (See "Idiopathic interstitial pneumonias: Classification and pathology".)
●Diffuse alveolar damage is usually associated with airspace consolidation or ground glass opacities in dependent locations.
●Findings suggestive of end-stage fibrosis include extensive reticular markings at the lung periphery, traction bronchiectasis, and honeycombing.
●Findings suggestive of nonspecific interstitial pneumonia include ground glass opacities, increased reticular markings in a subpleural location, and bronchiolectasis.
●Organizing pneumonia is manifest by ground glass opacities in a bilateral but asymmetric pattern or by airspace consolidation in a subpleural or peribronchial distribution. In the setting of bleomycin toxicity, organizing pneumonia may rarely present as one or more nodular densities that may mimic tumor metastases [108,110]. Often, the abnormalities are in a subpleural location.
●Hypersensitivity pneumonitis presents with diffuse, bilateral ground glass opacities and/or centrilobular nodules [109]. (See "High resolution computed tomography of the lungs".)
It is important not to confuse the new development of bleomycin-induced inflammatory nodules with progressive cancer. In difficult cases, biopsy may be needed. (See 'Diagnosis and differential diagnosis' below.)
Several reports describe increased uptake on 18-fluorodeoxyglucose-positron emission tomography (FDG-PET) scans in association with bleomycin toxicity [51,100-103]. The FDG uptake is described as diffuse, bilateral, and predominantly affecting the lower lobes and subpleural areas [51]. The FDG-PET findings can be subtle [102,103]; one study noted that lung FDG uptake is abnormal when it is equal to or greater than mediastinal uptake, as the lungs normally appear “cold” [102]. However, one problem with FDG-PET scans in this setting is that they are nonspecific and do not differentiate between bleomycin toxicity and infection.
Pulmonary function testing — We suggest obtaining pulmonary function tests (PFTs) in patients who develop dyspnea, cough, crackles on chest examination, or an abnormal chest radiograph while receiving bleomycin. PFTs, including spirometry, lung volumes, and DLCO, are helpful in identifying other common causes of cough and dyspnea such as asthma and chronic obstructive pulmonary disease (COPD). In patients with bleomycin-induced lung toxicity, spirometry and lung volumes typically demonstrate a restrictive pattern with decreases in forced vital capacity (FVC), total lung capacity (TLC), and functional residual capacity (FRC). In addition, the DLCO is usually decreased. For patients with a decline in DLCO of 25 percent or greater relative to baseline, we suggest omitting bleomycin. (See 'Permanent discontinuation of bleomycin' below.)
The majority of patients treated with bleomycin will have a decrease in their DLCO and those with significant pulmonary toxicity will have a decrease in lung volumes (eg, FVC and TLC). As an example, two randomized trials comparing a cisplatin, etoposide, plus bleomycin regimen versus cisplatin plus etoposide alone without bleomycin reported a median decline in DLCO of 14 to 20 percent in the bleomycin arms compared to 0 to 2 percent without bleomycin [32,36]. However, only a small percentage of patients exposed to bleomycin develop clinical signs or symptoms of lung toxicity.
Reductions in DLCO need to be interpreted with caution. When monitoring DLCO, a corrected “predicted value” for DLCO needs to be calculated for patients with anemia, as a decrease in hemoglobin leads to a reduction in carbon monoxide uptake (calculator 1 and calculator 2). The correction for anemia is discussed separately. (See "Diffusing capacity for carbon monoxide", section on 'Anemia'.)
Additionally, the DLCO can be affected by airflow limitation due to mediastinal masses, so the DLCO should only be considered valid if the inspiratory volume in the DLCO test is at least 85 percent of the vital capacity, as measured on spirometry. (See "Diffusing capacity for carbon monoxide", section on 'Quality of testing'.)
Bronchoalveolar lavage — The main role for bronchoalveolar lavage (BAL) in assessing possible bleomycin lung toxicity is to exclude infection or malignancy as causes of radiographic abnormalities. Data regarding typical BAL findings in bleomycin lung toxicity are limited. In one report, neutrophilia was most common, although eosinophilia may be seen in BAL samples from patients with hypersensitivity pneumonitis [3,111,112]. Cytologic atypia with hyperchromatic, multinucleated cells and prominent macronucleoli may also be seen [112].
Lung biopsy — Lung biopsy is rarely needed, as discontinuation of bleomycin usually leads to respiratory improvement. When needed, typically because the etiology of symptoms, signs, and/or radiographic abnormalities remains unclear after the above evaluation, tissue is obtained via video-assisted thoracoscopic surgery (VATS) or thoracotomy. (See "Role of lung biopsy in the diagnosis of interstitial lung disease", section on 'Overview'.)
Pathology — Gross lung specimens from subjects with bleomycin-induced lung injury typically demonstrate a subpleural distribution of lung injury and fibrosis [2]. Histopathology is nonspecific, although squamous metaplasia is a characteristic finding [4]. Various patterns of interstitial lung disease have been described, including end-stage fibrosis, nonspecific interstitial pneumonia, diffuse alveolar damage, organizing pneumonia, and hypersensitivity (eosinophilic) pneumonia [109]. More than one of these patterns may be present at the same time [104].
For patients with a pattern of diffuse alveolar damage, histopathologic findings include endothelial and type I epithelial cell necrosis, type II epithelial cell hyperplasia, and hyaline membranes. In some patients, the diffuse alveolar damage will resolve, in others it will progress to fibroproliferative lesions and excess collagen deposition.
Bleomycin-induced pulmonary nodules usually have histopathology similar to diffuse alveolar damage, although patterns of granulomatous infiltration, organizing pneumonia (previously known as bronchiolitis obliterans organizing pneumonia [BOOP]), and eosinophilia may occur [3,109,113]. (See "Idiopathic interstitial pneumonias: Classification and pathology" and "Cryptogenic organizing pneumonia".)
Hypersensitivity reactions demonstrate eosinophilic pneumonitis with focal consolidation [112,114,115]. Other inflammatory cells are also present, including plasma cells, lymphocytes, and mast cells. Focal areas of organizing pneumonia and dysplastic type II epithelial cells may be seen.
DIAGNOSIS AND DIFFERENTIAL DIAGNOSIS — The diagnosis of bleomycin-induced pulmonary toxicity is usually made based on the combination of a compatible clinical pattern or asymptomatic decline in the diffusing capacity for carbon monoxide (DLCO) by 25 percent during treatment with bleomycin, and the exclusion of infection or pulmonary involvement from the underlying malignancy.
Sputum analysis may be helpful to evaluate for the presence of infection, but often bronchoalveolar lavage (BAL) is performed for microbiologic and cytologic analysis. BAL findings in bleomycin-induced lung injury are nonspecific, so BAL is more helpful in the identification of infection or malignancy than in making a specific diagnosis of bleomycin-induced lung injury. (See 'Bronchoalveolar lavage' above and "Approach to the immunocompromised patient with fever and pulmonary infiltrates", section on 'Lung sampling'.)
When the diagnosis remains unclear, lung biopsy may be helpful, although histopathologic findings of drug-induced lung toxicity are nonspecific and require correlation with clinical, laboratory, and radiographic findings. (See "Approach to the immunocompromised patient with fever and pulmonary infiltrates" and "Role of lung biopsy in the diagnosis of interstitial lung disease".)
The differential diagnosis of bleomycin-induced lung injury includes lung infection, cardiogenic pulmonary edema, radiation-induced pulmonary fibrosis, metastatic disease, and an adverse reaction to other medications. (See "Pulmonary toxicity associated with systemic antineoplastic therapy: Clinical presentation, diagnosis, and treatment", section on 'Differential diagnosis'.)
TREATMENT — The optimal therapy for bleomycin-induced lung toxicity has not been established and no prospective trials of therapy have been performed. Based on clinical experience, the mainstay of treatment is prompt discontinuation of bleomycin for all patients with clinical signs or symptoms or an asymptomatic decrease in the diffusing capacity for carbon monoxide (DLCO). There is no widespread consensus as to the threshold of decline in DLCO that should prompt discontinuation of bleomycin, as discussed above. We use a 25 percent decrease in DLCO (adjusted for hemoglobin) from the pretreatment value. (See 'Screening asymptomatic patients for lung toxicity' above and 'Pulmonary function testing' above and "Diffusing capacity for carbon monoxide", section on 'Anemia'.)
Treatment with glucocorticoids is reserved for patients with symptomatic lung toxicity, as spontaneous resolution of asymptomatic radiographic opacities has been described [108].
Permanent discontinuation of bleomycin — The key step in the management of bleomycin-induced pulmonary toxicity is immediate and permanent discontinuation of bleomycin. Bleomycin should be discontinued in all patients with documented or strongly suspected bleomycin-induced lung injury, including an asymptomatic decline in DLCO of ≥25 percent. (See 'Screening asymptomatic patients for lung toxicity' above.)
Reinitiation of bleomycin is generally not recommended in patients with bleomycin-induced pulmonary toxicity and is contraindicated in patients with pulmonary fibrosis. Successful reinitiation of bleomycin in a patient with hypersensitivity pneumonitis related to bleomycin has been described in a couple of case reports [106,114]. However, we suggest not resuming bleomycin in patients who are thought to have hypersensitivity pneumonitis (as established by high resolution computed tomography and eosinophilia on bronchoalveolar lavage or histopathology) unless there is no equally effective alternative antineoplastic agent for the patient's disease.
Glucocorticoids for lung toxicity — There are no controlled trials of glucocorticoids or other immunosuppressive drugs for the treatment of bleomycin-induced lung toxicity, but based on case series, systemic glucocorticoids are usually initiated in patients with symptomatic respiratory impairment due to bleomycin toxicity, after exclusion of infection. It is not known whether patients with mild toxicity (eg, asymptomatic, no need for supplemental oxygen, decrease in DLCO as the only manifestation, minimal to no radiographic abnormalities) benefit from early introduction of glucocorticoids, or whether watchful waiting after discontinuation of bleomycin is acceptable. In these cases, we would stop the bleomycin and watch for worsening of symptoms and/or DLCO (adjusted for hemoglobin) before initiating glucocorticoids. (See 'Pulmonary function testing' above.)
The response to glucocorticoid therapy varies among the different patterns of pulmonary damage, as described in the following sections. However, the specific histologic type of lung toxicity (eg, nonspecific interstitial pneumonia, usual interstitial pneumonia, diffuse alveolar damage) is not definitively established in most patients, although organizing pneumonia and eosinophilic hypersensitivity pneumonitis may be strongly suggested based on the radiographic pattern or presence of peripheral blood or bronchoalveolar lavage eosinophilia. Thus, the decision to initiate glucocorticoid therapy is usually made without knowing the exact histopathologic pattern of pulmonary injury.
●Interstitial pneumonitis and diffuse alveolar damage – Case reports and case series have described substantial recovery in patients with subacute or acute bleomycin-induced interstitial pneumonitis when a significant inflammatory pneumonitis was present (eg, nonspecific interstitial pneumonia or acute interstitial pneumonia [diffuse alveolar damage]) [2,104,116,117]. In a series of 10 patients with bleomycin-induced lung toxicity, the seven patients who were treated with systemic glucocorticoids experienced significant clinical and radiographic improvements, while the three patients who were not treated with glucocorticoids died [117].
The optimal dosing and duration of glucocorticoid therapy for bleomycin-induced lung injury are not known. Based on clinical experience and data from case series, we typically initiate treatment with prednisone at 0.75 to 1 mg/kg (using ideal body weight) per day up to 100 mg per day [117]. After four to six weeks the dose of prednisone is tapered gradually over an additional four to six months, in accordance with the patient's condition and clinical response.
Short-term improvement occurs in 50 to 70 percent of glucocorticoid-treated patients, but symptoms may relapse when therapy is tapered [117]. Some patients who respond to glucocorticoid therapy initially may subsequently develop progressive interstitial pulmonary fibrosis that is unresponsive to glucocorticoid therapy. Pulmonary function abnormalities may or may not improve with therapy and may recur after a period of one to five years, even in patients with an initially positive response to glucocorticoids.
●Interstitial fibrosis – Some patients who have a later, more indolent onset of interstitial lung disease after bleomycin therapy may have a clinical presentation suggestive of usual interstitial pneumonia with progressive fibrosis and less active inflammation. In these patients, the radiographic pattern on high-resolution CT is consistent with usual interstitial pneumonia (eg, increased reticular markings and evidence of honeycombing, but minimal or absent ground glass opacities). These patients may not have glucocorticoid-responsive disease. The decision to initiate systemic glucocorticoids must take into account the low likelihood of benefit and the substantial adverse effects of glucocorticoids. If a clear response to systemic glucocorticoids does not occur within four to six weeks, we discontinue glucocorticoids to avoid side effects related to long-term glucocorticoid therapy. (See 'Imaging' above and "Idiopathic interstitial pneumonias: Classification and pathology", section on 'Usual interstitial pneumonia'.)
●Organizing pneumonia – We recommend that patients with symptomatic bleomycin-induced organizing pneumonia be treated with systemic glucocorticoids, although occasionally, organizing pneumonia will remit without therapy. We usually follow the glucocorticoid dosing used for cryptogenic organizing pneumonia. The initial dose of prednisone is usually 0.75 to 1 mg/kg per day (using ideal body weight) to a maximum of 100 mg/day given as a single oral dose in the morning. Treatment of cryptogenic organizing pneumonia is discussed separately. (See "Cryptogenic organizing pneumonia", section on 'Treatment'.)
●Hypersensitivity pneumonitis – For patients who are diagnosed with hypersensitivity pneumonitis based on an acute onset of pneumonitis, a radiographic appearance consistent with hypersensitivity pneumonitis, and either bronchoalveolar lavage eosinophilia or histopathologic confirmation, we typically initiate systemic glucocorticoids, after infection is excluded.
There are no data regarding the optimal dosing of prednisone for bleomycin-induced hypersensitivity pneumonitis, but the response is usually excellent, as long as bleomycin is discontinued. A reasonable choice is to start with 0.75 to 1 mg/kg (based on ideal body weight) per day, to a maximum of 100 mg/day, for the first four to six weeks. In comparison with other forms of bleomycin-induced lung toxicity, there is usually a more rapid response to glucocorticoid treatment, although this is based on anecdotal reports. Tapering of prednisone is usually accomplished over a few months [114].
Supplemental oxygen and future perioperative management — Although the evidence is largely anecdotal and inconsistent, exposure to high inspired oxygen concentrations, even many years following exposure to bleomycin, may increase the risk for pulmonary toxicity. (See 'High fraction of inspired oxygen' above.)
For patients with prior bleomycin exposure who have hypoxemia, supplemental oxygen is administered sparingly to achieve an oxygen saturation of 89 to 92 percent. The priority is to maintain adequate oxygen saturation, even if that requires a high fraction of inspired oxygen (FiO2).
For bleomycin-exposed patients who are undergoing surgery, intravenous fluids are administered sparingly to avoid volume overload and perioperative pulmonary edema [45,68,118].
SUMMARY AND RECOMMENDATIONS
●Pathogenesis – There are four main types of pulmonary toxicity associated with bleomycin: subacute progressive pulmonary fibrosis, hypersensitivity pneumonitis, organizing pneumonia, and an acute chest pain syndrome during rapid infusion. (See 'Pharmacology and pathogenesis of lung injury' above and 'Clinical presentation' above.)
●Risk factors
•The risk of bleomycin-induced lung toxicity appears to be higher in older patients and those with renal insufficiency. (See 'Risk factors' above.)
•Administration of higher bleomycin doses (>400 units) clearly increases the risk of lung injury, although injury can occur at doses <50 units.
•Thoracic irradiation and concurrent administration of cisplatin at high doses may increase the risk of pulmonary toxicity.
●Screening asymptomatic patients
•Although practice is variable, we suggest assessment of pulmonary function tests (PFTs), including spirometry and diffusing capacity for carbon monoxide (DLCO), at baseline and at intervals during therapy for most adults with germ cell tumors and older adults with Hodgkin lymphoma who are receiving a bleomycin-containing chemotherapy regimen.
The optimal frequency of testing is not established. For patients with germ cell tumors, we monitor PFTs prior to each new treatment cycle. For older adults with Hodgkin disease receiving a bleomycin-containing regimen such as ABVD (doxorubicin, bleomycin, vinblastine, dacarbazine), we assess PFTs after two to four cycles of therapy. (See 'Screening asymptomatic patients for lung toxicity' above.)
•We do not generally screen asymptomatic patients with either chest radiographs or fluorodeoxyglucose positron emission tomography scanning (FDG-PET). However, for patients receiving ABVD for Hodgkin lymphoma who undergo interval FDG-PET for early response assessment, we discontinue bleomycin if the interim scan shows increased FDG uptake in the lungs suggestive of bleomycin lung toxicity. (See 'Role of FDG-PET' above.)
●Clinical presentation and evaluation of symptomatic patients
•Clinical manifestations usually develop subacutely between one and six months after treatment initiation, but may occur more than six months after treatment discontinuation. Symptoms and signs include nonproductive cough, dyspnea, pleuritic or substernal chest pain, fever, tachypnea, crackles, lung restriction, and hypoxemia. (See 'Clinical presentation' above.)
•The classic radiographic pattern of bleomycin-induced pulmonary fibrosis includes bibasilar subpleural opacities with volume loss and blunting of the costophrenic angles; however, fine nodular densities may also be seen. Honeycombing may develop, as the fibrosis progresses. (See 'Imaging' above.)
•The main role for bronchoalveolar lavage is to exclude infection or malignancy as causes of radiographic abnormalities. (See 'Bronchoalveolar lavage' above.)
•Lung biopsy is rarely needed, as bleomycin discontinuation usually leads to respiratory improvement. (See 'Lung biopsy' above.)
●Treatment
•We recommend permanently discontinuing bleomycin therapy in patients with documented or strongly suspected bleomycin-induced lung injury, including an asymptomatic decrease in the DLCO of 25 percent or more (Grade 1A). We do not generally reintroduce bleomycin in these patients unless the goal of therapy is cure and there is no equally effective alternative antineoplastic agent for the patient's disease. (See 'Permanent discontinuation of bleomycin' above.)
•For patients with symptomatic acute or subacute bleomycin-induced pulmonary toxicity and impairment on PFTs, we recommend systemic glucocorticoids (Grade 1B). For patients with mild toxicity (eg, asymptomatic, no need for supplemental oxygen, decrease in DLCO as the only manifestation, minimal to no radiographic abnormalities), we would stop the bleomycin and watch for worsening of symptoms and/or DLCO before initiating glucocorticoids. (See 'Glucocorticoids for lung toxicity' above.)
•For patients with a clinical presentation of chronic pulmonary fibrosis (late, indolent onset and CT scan consistent with usual interstitial pneumonia), we suggest not administering systemic glucocorticoids based on the lack of benefit in patients with idiopathic pulmonary fibrosis (Grade 2C). (See 'Treatment' above.)
●Future perioperative management – Although the evidence is anecdotal as to whether administration of high inspired fractions of oxygen may provoke or exacerbate pulmonary toxicity several years after treatment with bleomycin, we suggest carefully titrating any supplemental oxygen to provide only enough oxygen to maintain the oxygen saturation at 89 to 92 percent (Grade 2C). (See 'Supplemental oxygen and future perioperative management' above.)
For bleomycin-exposed patients who undergo subsequent surgery, intravenous fluids should be administered sparingly to avoid volume overload and perioperative pulmonary edema.
ACKNOWLEDGMENTS — The editorial staff at UpToDate acknowledge Philip W Kantoff, MD, Sumanta Pal, MD, and Nicholas Vander Els, MD, who contributed to earlier versions of this topic review.
