Incidence of Radiation Lung Injury and Predictive Factors

The most important factor influencing development of clinically relevant radiation lung injury is the volume of lung irradiated; this issue is extensively discussed in the radiation oncology literature.11–15 In radiation oncology, the lung is considered a parallel-architecture organ, meaning that destruction of very small portions of it should not cause overall organ dysfunction. This belief is supported by the fact that wedge resection of a small portion of lung tissue or stereotactic radiosurgery to a very small lung tumor is well tolerated by most patients. In contrast, other organs, such as the spinal cord, are considered series-architecture organs, in which destruction of one region will lead to irreversible dysfunction downstream from that injury.¹⁶ The series-architecture model of radiation-induced toxicity does apply to other tissues within the lung, such as major bronchi and/or pulmonary vessels, where the result of injury to a small volume can be catastrophic (e.g., fistula and massive hemorrhage).¹⁷

The percentage of lung irradiated is often expressed by radiation oncologists as the “Vx,” where x = a certain dose in Gy. V20 indicates the percentage of a patient’s total lung volume that was irradiated to a dose of at least 20 Gy. Mean lung dose (expressed in Gy) is another parameter commonly used to express the extent to which a patient’s lung volume has been irradiated. Most clinical trials for locally advanced lung cancer require that the V20 be under a certain threshold, such as 35%. However, no single parameter can express the total radiation exposure, which is best described by a two-dimensional graphical curve called a dose-volume histogram (Fig. 58-2 provides an example).

Irradiation of the entire lung volume (bilateral lungs) is uncommon today, except as part of total body irradiation for selected BMT conditioning regimens. As reviewed by Sampath and colleagues,¹⁸ the therapeutic index for whole-lung irradiation is extraordinarily narrow and highly dependent on total dose, fractionation, partial lung transmission shielding, and dose rate. A typical dose of 12 Gy total body irradiation has an approximately 11% rate of severe pneumonitis, compared with approximately 2% when lung transmission shielding is used to reduce the lung dose to 6 Gy.

At the other extreme, irradiation of small lung volumes, as with stereotactic body radiation therapy (SBRT, also known as radiosurgery) for early-stage non–small cell lung cancer (NSCLC) rarely results in high-grade RP, despite its use in a compromised patient population.21–21 The field of SBRT for lung cancer is evolving very quickly, and it is likely that with longer follow-up of patients, a greater understanding of the similarities and differences from conventional external beam radiation therapy (XRT) with regard to toxicity will be achieved.

In tangential irradiation of the breast (after lumpectomy) or chest wall (after mastectomy), the risk of clinically significant RP is approximately 1%, increasing to approximately 5% with the addition of irradiation of the regional (axillary/supraclavicular) nodes.²² An association between dose-volume irradiated and pneumonitis risk has also been shown in patients treated for esophageal carcinoma.²³

Level I evidence of a strong relationship between pneumonopathy and the volume irradiated comes from a randomized trial from China.²⁴ This study showed that the incidence of RP was significantly reduced from 29% to 17% with the use of “involved-field” radiotherapy, despite the use of higher tumor radiotherapy doses in the involved field arm.

Understanding the true incidence of radiation pneumonopathy is complicated by multiple factors, including ambiguities in scoring,²⁶ uncertainties related to concurrent medical conditions, and limitations of the historical and current standards for defining and grading this illness.²⁵ Kocak and colleagues²⁵ found that 28% of their suspected RP had confounding medical conditions that made the clinical diagnosis uncertain. Yirmibesoglu and colleagues²⁶ concluded that 48% of RP cases were “hard to score” (vs. “unambiguous”) because of confounding factors (e.g., tumor progression, acute exacerbation of COPD, and infection).

The traditional scoring system developed by the Radiation Therapy Oncology Group (RTOG), which dates back to the early 1970s, is shown in Table 58-1.²⁷ This system is no longer generally used because of limitations such as an arbitrary cutoff of 90 days between acute and late radiation lung injury²⁸; however, understanding of this system is important because numerous publications rely on it. Modern toxicity scoring is performed according to the Common Terminology Criteria (CTC), currently at version 4 (CTCv4).²⁹ The CTCv4 definitions and grading for select pulmonary events are shown in Table 58-2. A comparison of CTCv3 or the old RTOG criteria to CTCv4 shows that numerous changes have been made to the scoring criteria in the pulmonology section. Separate scales are no longer used for early and late radiation therapy-related events.

Tucker et al.³⁰ found that the scoring system used (CTCv2 vs. CTCv3 vs. RTOG) greatly affected the results of analyses of the risk of RP and the radiation dose-volume factors associated with RP.

Although radiation dose-volume parameters are the major predictor of radiation pneumonopathy, the addition of systemic therapy must also be considered. It appears that neoadjuvant chemotherapy before radiotherapy does not greatly affect the risk of radiation pneumonopathy. However, concurrent chemoradiotherapy probably increases the risk significantly; the maximum tolerated dose of localized radiotherapy with concurrent chemotherapy for NSCLC is generally considered to be 74 Gy,³¹ whereas the maximum tolerated dose for radiotherapy alone is usually considered to be higher than that.¹⁴

A combination of concurrent radiotherapy and gemcitabine appears to have a particularly high rate of radiation pneumonopathy, and extreme caution is advised for this combination (Table 58-3).³² Several other drugs, most notably the anthracyclines (e.g., doxorubicin), methotrexate, and bleomycin, should be considered contraindicated during thoracic radiotherapy.³³ Concurrent thoracic radiotherapy plus a taxane (paclitaxel or docetaxel) is a commonly used and generally safe combination, although it has been suggested that the risk of radiation pneumonopathy might be slightly higher than expected from historical series based on older chemoradiotherapy regimens, particularly with weekly taxane schedules.^36–36 It may be especially prudent to use three-dimensional radiotherapy techniques and to carefully analyze lung dose-volume histogram parameters in patients who are receiving a concurrent taxane.

Fewer data exist regarding other systemic agents, including biological agents. These agents are discussed briefly in Part II of this chapter.

Nontreatment factors that appear to be predictive of radiation lung injury have been studied. Not surprisingly, pretreatment performance status has been shown to correlate with development of clinical radiation pneumonopathy.³⁷ Underlying pretreatment pulmonary function is probably an important factor as well,³⁸ although study results have been inconsistent,^39,40 perhaps because in many cases, poor pretreatment pulmonary function is due to effects of the cancer itself; an effective treatment (such as radiotherapy) may then substantially improve rather than harm pulmonary function. Other possible risk factors for toxicity include prior thoracic irradiation, volume loss due to lung collapse, forced expiratory volume in one second <2 L for severe (grade ≥3) RP,³⁷ older age,³³ female gender,^21,33 never or current nonsmokers^33,41,42 and poor pretreatment performance status, and pretreatment COPD in patients with lung cancer.⁴³ Risk factors significantly associated with RP in patients with Hodgkin lymphoma were minimum lethal dose ≥13.5 Gy and V20 ≥33.5% when initial combined modality therapy consisted of Adriamycin (doxorubicin), bleomycin, vinblastine, and dacarbazine and mediastinal radiotherapy.⁴⁴ Use of angiotensin-converting enzyme (ACE) inhibitors may be associated with a lower risk of RP,⁴⁵ although this is controversial and not confirmed by any prospective data.

Risk of radiation-induced lung injury may be higher in patients with certain genetic variations. Studies have explored ataxia-telangiectasia mutated (ATM) gene⁴⁶ or TGFβ1 gene polymorphisms⁴⁷ as they relate to radiation lung injury, although this parameter is not yet used clinically.

Diagnosis and Management of (Acute and Subacute) Radiation Pneumonitis

With modern, three-dimensional conformal, multifield radiation therapy, SBRT, or intensity-modulated radiation therapy, it is not possible to simply look for pathognomonic rectangular infiltrates on a chest radiograph to confirm RP. A patient who has undergone radiation treatment (particularly if a large volume of lung was irradiated to a dose >20 Gy) and has symptoms should undergo computed tomography (CT) scanning, in some cases with high-resolution CT imaging. CT angiography may also be indicated to rule out a pulmonary embolism. Imaging will typically reveal an interstitial infiltrate and/or ground-glass appearance, which can be very difficult to distinguish from an infection or tumor-related changes. Suspected moderate to severe radiation pneumonopathy should be evaluated by a pulmonologist for consideration of bronchoscopy to rule out infection, particularly if fever is present.⁴⁸

Pulse oximetry or arterial blood gas testing should be performed to assess the need for supplemental oxygen. Pulmonary function testing, especially the diffusion capacity of the lung for carbon monoxide (DLCO) test, is useful as part of the diagnostic workup, in particular to assess the severity of gas-exchange dysfunction, although pulmonary function tests are not strongly correlated to clinical symptoms of radiation-induced lung disease.³³ Spirometric parameters of pulmonary function are frequently reversible after radiation pneumonopathy, whereas DLCO abnormalities are less likely to improve.⁴⁹ The tests and procedures to be considered in the evaluation of patients with cancer who have suspected pneumonopathy are summarized in Table 58-4. This workup should be considered appropriate for either suspected radiation-related or chemotherapy-related pneumonopathy.

If the clinical manifestations and test results are consistent with grade 2 or greater RP, administration of corticosteroids should be instituted in most cases. Controlled randomized trials of corticosteroids for RP have not been conducted with human subjects. However, the efficacy of corticosteroids has been well established in nonrandomized clinical studies⁵⁰ and in preclinical models.⁵¹ No single “standard” dose schedule exists for steroid therapy for persons with RP; the exact schedule must be tailored to the individual patient. In general, for severe (grade 3) RP, prednisone, approximately 1 mg/kg/day, is indicated for 2 weeks. Most clinicians suggest starting with 60 mg/day prednisone, then tapering as symptoms improve. Brief hospitalization for intravenous administration of steroids may be indicated. Early onset of RP after the completion of radiotherapy may predict a more virulent course and therefore might require a more aggressive management approach.⁷ Moderate RP (grade 2) may be effectively managed with somewhat lower initial doses of steroids (e.g., 0.5 to 0.75 mg/kg/day of prednisone); however, the patient must be evaluated frequently to ascertain that his or her condition is not progressing to grade 3 or worse RP. After several weeks, the dose should be tapered gently, by approximately 10 mg every 2 weeks. It is not unusual for patients to have a symptomatic relapse in the setting of steroid taper.⁵² If relapse occurs, it is important to rule out concomitant infection. If the diagnosis of recurrent RP is confirmed, the steroid dose should be increased and titrated accordingly.

With these guidelines, the typical patient with grade 3 or very intense grade 2 radiation lung injury may require steroids for approximately up to 4 months. Some patients need steroids for considerably longer, although the benefit after 6 months is dubious (when RP has generally resolved and might be superseded by fibrosis). At the outset, the physician should explain to the patient the potential need and implications of longer-term steroid use. A proton-pump inhibitor or histamine₂ blocker should be prescribed to counteract gastritis. Consideration should be given to evaluation for, and medical prophylaxis against, osteoporosis. Patients should be counseled about exercise (as tolerated by their pulmonary symptoms) and diet to minimize problems with steroid-induced hyperglycemia and muscle wasting. Blood chemistry values, including fasting glucose, liver function tests, and albumin, should be checked periodically. If a diuretic is being used with steroids, it is important to check serum electrolyte levels frequently. Adrenal insufficiency should also be considered when steroids are tapered, and some patients may require long-term or even lifelong maintenance steroids.

It is uncertain whether a low-dose “prophylactic” antibiotic should be prescribed. In any given patient with interstitial pneumonitis after thoracic radiotherapy, it can be very difficult to entirely rule out a concomitant infection, particularly if bronchoscopy or another invasive diagnostic procedure was not performed.⁴⁸ In light of the profound lymphopenia that many patients have after chemoradiation therapy and steroid treatment (as with typical concurrent paclitaxel-radiation therapy regimens), it might be appropriate to prescribe an every-other-day dose of trimethoprim-sulfamethoxazole (Bactrim) when starting high-dose steroids.^53,54 If findings of a chest CT image suggest the presence of co-existent active infection, broader and more intense antibiotics are indicated. The patient should undergo bronchoscopy if no improvement occurs after several days of steroid and antibiotic therapy.

The prognosis of grade 1 to 2 RP is relatively good with meticulous supportive care and the use of steroids as needed. Grade 3 RP, however, at least in patients with lung cancer, has a much worse prognosis.7,13 It is uncertain whether this worse prognosis is the direct result of RP or coexisting problems, including tumor recurrence and infection.

Management of radiation pneumonopathy that is refractory to steroids and/or that occurs in a patient who has severe contraindications to steroids is uncertain. The literature includes case reports of the use of antirheumatic medications such as cyclosporin,55,56 but little scientific evidence supports their use.

As described in the following sections, patients with RP should receive supportive care based on careful and frequent assessment of their symptoms and risk factors for further complications of their illness.

Management of (Chronic and Late) Radiation Pulmonary Fibrosis

Radiation pulmonary fibrosis, unlike RP, is not relieved with steroids. Some patients undergoing steroid therapy for RP eventually start to experience worsening pulmonary function. This decline may be caused by pulmonary fibrosis but also may be caused at least in part by disorders such as infection, cardiac problems, and pulmonary embolism, and reevaluation is indicated. If it has been more than 6 months since the patient has undergone radiation therapy and/or chemotherapy, increasing the steroid dose is unlikely to yield benefit.

Management of radiation pulmonary fibrosis is supportive. Emphasis is on administration of oxygen, management of acute infection, bronchodilator therapy if needed, gentle diuresis if needed, and maximization of the other components of tissue oxygen delivery (e.g., cardiac and blood pressure medications and correction of anemia). Acute episodes of RP may still occur, particularly if the patient receives a drug that can induce a “radiation recall” reaction (e.g., gemcitabine and taxanes). Short courses of steroids might be needed periodically. It is unknown whether pulmonary rehabilitation programs involving exercise and weight maintenance are helpful for radiation pulmonary fibrosis. Patients who are disabled or immobilized because of their pulmonary insufficiency might benefit from deep venous thrombosis prophylaxis. Use of tobacco should be strictly avoided.

Trials of Prevention or Management of Radiation Pulmonary Fibrosis

Until the 1990s, radiation-induced lung injury was considered inevitable and untreatable, the result of delayed mitotic death of parenchymal or vascular cells.⁵⁷ As understanding of the mechanisms of lung injury has improved, prophylactic treatments have been studied, but few pharmacologic agents are approved for prevention of radiation-induced lung injury. Three main classes of radiation countermeasure agents are (1) radioprotective agents, given before irradiation; (2) radiation mitigators, given after irradiation but before expression of tissue damage; and (3) therapeutic agents, given after irradiation as palliation or support.⁵⁸

Only one radioprotective agent, a synthetic thiol-reducing agent called amifostine, has been approved by the U.S. Food and Drug Administration to protect normal tissues from radiation during treatment, and this was to prevent renal and salivary gland damage, not specifically for pulmonary toxicity.⁵⁸ The efficacy of amifostine to prevent RP is uncertain; results of randomized trials have been mixed.^61–61 A relatively large confirmatory study found no significant differences.⁶² Amifostine is logistically and financially difficult to deliver and is associated with its own toxicities (e.g., nausea, fatigue, skin rash, hypotension, and syncope). Its effectiveness is likely highly correlated with dose and schedule, and amifostine has no known role or rationale as a treatment once radiation has been completed.

Other nonspecific antioxidants have been studied as means of decreasing radiation fibrosis, although they are not necessarily specific for the lung.63–66 Multiple dietary^69–69 plant⁷⁰ antioxidants have been studied in vitro and in animal models to prevent radiation lung damage. Gene therapy to deliver antioxidants to the lung is under study. Researchers from the University of Pittsburgh have developed a system in which the gene for the potent antioxidant enzyme manganese superoxide dismutase can be transfected into a plasmid and administered by inhalation.⁶⁴ This approach, which is still highly experimental, might offer a preventive and/or therapeutic measure.

The use of angiotensin-converting enzyme (ACE) inhibitors, particularly captopril, showed promise in preclinical animal studies.⁷¹ In one retrospective analysis of patients with stage I-III lung cancer who were treated with definitive radiation, the incidental concurrent use of ACE inhibitors was associated with less radiation-induced pulmonary toxicity.⁷² This finding has not been validated in a prospective or controlled trial.⁷³ The mechanism of efficacy is unclear; it might be related to the beneficial vasodilatory effects of these drugs on small blood vessels or attributed to antioxidant activity.

Two modest-sized randomized trials from Turkey suggested that pentoxifylline (PTX) may help prevent acute and late lung tissue damage.74,75 In one study, PTX or placebo was given at a dose of 400 mg three times a day throughout radiotherapy; results showed significantly better DLCO and perfusion scan results at 3- and 6-month follow-up among patients randomly assigned to PTX. In the other study, patients were randomized to control versus PTX (400 mg three times a day during radiotherapy and 400 mg daily after radiotherapy for 3 months) plus vitamin E (300 mg twice daily during radiotherapy and 300 mg daily after radiotherapy). Results showed less RP in the arm randomized to PTX/vitamin E. The potential mechanism by which PTX ameliorates RP is unknown, and these studies have not been validated in other randomized trials. No meaningful data are available regarding the efficacy of PTX as a treatment (as opposed to preventative agent) for radiation pneumonopathy.

Increased understanding of the cytokine-based mechanisms of radiation lung injury may offer opportunities for intervention.65,76 Because TGF-β is thought to be the dominant profibrotic cytokine and perhaps even a cause of RP (particularly late radiation pulmonary fibrosis), attempts are underway to develop molecules with anti-TGF-β activity. At this time, no molecularly targeted treatments for established radiation lung injury (RP or radiation pulmonary fibrosis) are clinically available. Steroids (for RP at least) remain the only known treatment.

Bleomycin

Bleomycin is the chemotherapy drug most commonly associated with lung damage. The reported incidence ranges from 3% to 40%.77,78 This drug is used predominantly in the management of Hodgkin disease and germ cell tumors, cancers that occur mainly in younger patients with less underlying pulmonary comorbidity than patients with lung cancer. Bleomycin pneumonopathy and RP have a number of similarities, including two patterns of disease (pneumonitis and fibrosis). As with radiation, the clinical manifestations of bleomycin lung toxicity usually occur weeks to months after the initiation of treatment. The infiltrates can be diffuse or limited to basilar and subpleural aspects of the lungs.⁷⁹ A nodular pattern occasionally occurs, mimicking cancer progression. Imaging other than CT scanning is investigational, although there are reports of abnormal fluorodeoxyglucose-positron emission tomography scan results in persons with bleomycin pneumonopathy.⁸⁰

As in radiation pneumonopathy, dyspnea is the primary symptom of bleomycin lung toxicity, although, as is the case with radiotherapy, other symptoms such as cough and fever often occur as well. Pulmonary function testing shows a restrictive pattern; DLCO is frequently abnormal. DLCO should be measured before bleomycin is started and periodically between cycles. A significant decrease in DLCO should prompt consideration of discontinuation of this drug. Although no consensus exists on what represents a “significant decrease” in DLCO, some investigators have conservatively used a criterion of a 20% decrease to warrant discontinuation of bleomycin.⁸¹

An association appears to exist between cumulative bleomycin dose and risk of pneumonopathy. Again, no consensus exists regarding the absolute maximum cumulative bleomycin dose for an individual patient; values of 300 to 400 mg maximum have been suggested.⁷⁷ (A typical cumulative dose of bleomycin from six cycles of Adriamycin [doxorubicin], bleomycin, vinblastine, and dacarbazine [used to treat persons with Hodgkin disease] is approximately 120 mg/m²). Again, however, it must be stressed that occasionally life-threatening or even fatal pneumonopathy can occur with cumulative bleomycin doses less than 100 mg. Another important predictive factor (in addition to age and dose) for bleomycin pneumonopathy appears to be renal insufficiency,^77,82 which is particularly concerning in patients who are receiving nephrotoxic agents such as cisplatin. Other possible risk factors may include concomitant use of a colony-stimulating factor,⁸³ combined treatment with radiotherapy, and concomitant delivery with gemcitabine.^84,85 Exposure to high oxygen concentrations as part of operative anesthesia has been associated with potentiation of bleomycin pneumonopathy, particularly in animal models,⁸⁶ and in some cases led to acute respiratory distress syndrome.^87,88

Extreme caution is advised when supplemental oxygen is being electively prescribed for patients treated with bleomycin, especially those who are currently undergoing or recently finished bleomycin (within less than 12 months) and/or have documented bleomycin pneumonopathy.⁸⁹

Bleomycin pneumonitis is rarely fatal; one study estimated a lethality risk of about 3%, with a suggestion that older age was associated with a higher risk for death.⁹⁰ Most patients achieve complete or near-complete recovery.^90–94 Corticosteroids may have some benefit,^77,91,95 although the data are not as clear as for radiation pneumonitis. Some cases of steroid-responsive bleomycin pneumonitis/fibrosis may represent early hypersensitivity, bronchiolitis obliterans with organizing pneumonia–like events. After bleomycin chemotherapy, cancer survivors may have significant declines in pulmonary function for approximately 6 months, but by 2 years after chemotherapy, few have significant respiratory dysfunction.

Antimetabolites

Methotrexate, which is commonly used in therapy for rheumatoid arthritis and for cancer, is the antimetabolite drug that is most commonly linked to pulmonary injury. It can cause interstitial pneumonitis similar to that caused by bleomycin and other drugs.96–99 Steroid therapy may be indicated for severe cases.

Gemcitabine has become one of the most commonly used chemotherapy drugs for solid tumors, including lung cancer. When gemcitabine is used as a single agent in previously untreated patients with cancer, the risk of significant pulmonary toxicity is small, although it is associated with an occasional severe respiratory insufficiency that may be related to capillary leak syndrome.¹⁰⁰ However, increasing data suggest that combinations of gemcitabine with other agents may severely potentiate those agents’ pneumonopathy (see the discussion in previous sections regarding thoracic radiotherapy and bleomycin)—in particular, the combination of gemcitabine with taxanes.^101,102 Steroid therapy can be a useful adjunct to supportive treatment.⁹² Gemcitabine may also be associated with radiation recall pneumonitis, even a year or more after thoracic radiation therapy. Gemcitabine radiation recall reaction appears to be more likely to affect parenchymal organs such as the lung than the more common dermatologic radiation recall reactions associated with other chemotherapeutic agents.⁹³

Fludarabine is now commonly used to manage hematologic malignant disease, including chronic lymphocytic leukemia, and, like other antimetabolite drugs, it is relatively well tolerated. In approximately 8% of cases, fludarabine causes pulmonary toxicity characterized by fever and interstitial pneumonitis.94,101 Although this syndrome is probably steroid responsive,¹⁰³ it is critical to rule out opportunistic infection given the patient population usually treated with this drug.

Alkylators and Nitrosoureas

Alkylators and nitrosoureas are frequent components of conditioning regimens for BMT, a procedure that is associated with a high rate of pulmonary toxicity. It is difficult to isolate the effect of these drugs in this setting, in which many other insults to the lung often occur, but it is likely that these drugs contribute to pulmonary toxicity, particularly at BMT dose intensity. Results suggest that pulmonary toxicity from BMT with regimens including whole-body irradiation is similar to that from BMT with regimens that do not include whole-body irradiation,104,105 which underscores the relationship between high-dose alkylator therapy and pneumonopathy.

In the non-BMT setting, classic alkylators, such as cyclophosphamide, ifosfamide, and melphalan, have occasionally been associated with interstitial pneumonitis and acute or subacute dyspnea that may be steroid responsive.106,107 However, these drugs have caused long-term pulmonary fibrosis, even several decades after treatment.¹⁰⁶

For nitrosoureas such as BCNU, pulmonary injury is considered a dose-limiting toxicity and has been well studied.¹⁰⁸ The predominant problem with these agents is pulmonary fibrosis, which appears to be chronic and dependent on dose and patient age.^111–111 In one study, BCNU pulmonary fibrosis in patients treated for glioma was rare at cumulative doses less than 960 mg/m².¹¹¹ The association between other nitrosoureas and pulmonary toxicity is not commonly reported but does occasionally occur.^112,113 The pulmonary fibrosis that is caused by nitrosoureas does not appear to be steroid responsive. The unpredictable and untreatable nature of nitrosourea-induced pneumonopathy has recently limited the use of these drugs in several settings.¹¹³

Anthracyclines and Other Antitumor Agents

Essentially every cytotoxic chemotherapy has been associated with pulmonary toxicity. Case reports and/or literature reviews can be identified even for drugs that are rarely associated with pneumonopathy, such as vinorelbine,¹¹⁴ paclitaxel,¹¹⁵ and oxaliplatin.¹¹⁶ Response to steroids has been variable. Several other agents, however, deserve special mention.

The antibiotic chemotherapy drug mitomycin is well known as a potentially pneumotoxic agent that can cause noncardiogenic pulmonary edema, pneumonitis, and pleural effusion. Toxicity is difficult to predict and not clearly dose-related but is estimated to occur in 5% of patients or more.117,118 In a prospective study of patients treated with mitomycin, 28% of patients tested had significant (>20%) declines in DLCO, and 5% had grade 3+ lung toxicity.¹¹⁸ Mitomycin-induced pulmonary toxicity appears to be steroid responsive.¹¹⁹

Anthracyclines are vital drugs for the treatment of breast cancer among other malignancies; although they are associated with significant cardiac risk (which can mimic pulmonary toxicity), they rarely independently cause pneumonopathy. However, the combination of anthracyclines such as doxorubicin with thoracic radiotherapy can be highly toxic. For example, a randomized trial of concurrent versus sequential chemoradiotherapy (including doxorubicin) for small cell lung carcinoma was closed early because six fatal cases of pneumonopathy developed among 82 patients who were treated concurrently.¹²⁰ There may be a risk of radiation recall-like reactions in patients receiving doxorubicin after high-dose thoracic radiation therapy, at least in the vulnerable lung cancer population.¹²¹

Like the anthracyclines, the taxanes (paclitaxel and docetaxel) are widely used solid tumor antineoplastic drugs. Docetaxel seems to be more associated with pneumonopathy than does paclitaxel. The risk of pneumonitis appears to be schedule-related, with a significantly higher risk with the use of weekly (versus every 3 weeks) docetaxel.¹⁰⁷ This finding contrasts with the nonpulmonary toxicities of this drug, which appear to be lessened with the weekly schedule.¹²² Docetaxel has also been associated with enhancement of radiation pneumonopathy in patients with lung cancer.¹²³ The combination of paclitaxel plus radiotherapy after anthracycline-based chemotherapy might not be feasible in patients with breast cancer because of pulmonary toxicity¹²⁴ despite the relatively modest amounts of lung in the irradiated field. As was noted previously, the combination of taxanes with gemcitabine appears to result in a significant risk of pneumonopathy⁹³ and should be used with extreme caution in patients with underlying lung disease.

Biological Agents

Increasingly, cytotoxic therapies such as chemotherapy and/or radiotherapy are being replaced or supplemented by “biological” agents that kill or inhibit cancer cells through highly specific pathways independent of direct DNA damage. Many of these drugs are still very new, and their toxicity profiles are still being generated. Therefore data regarding pulmonary toxicity of these agents are incomplete.

Immunomodulatory anticancer agents are one of the oldest classes of biological therapies. Pneumonopathy associated with interleukins or interferons occasionally occurs,¹²⁵ as well as a syndrome of noncardiogenic pulmonary edema, which differs from typical drug-induced interstitial pneumonitis.¹²⁶ Noncardiogenic pulmonary edema is generally reversible with steroids and supportive care and rarely progresses to fibrosis.

Several randomized trials suggested that the addition of interferon to conventional lung cancer therapy increased the risk of pneumonopathy.127,128

Targeted therapies against signal transduction pathways have become a heralded advance in several hematologic and solid tumor types. Most of these drugs have not been studied extensively enough to fully characterize their pulmonary risks. An overview of incidence and specific patterns of pulmonary toxicity seen with molecularly targeted agents used for cancer therapy follows (see Table 58-6).

Antiepidermal Growth Factor Receptor Tyrosine Kinase Inhibitors

Gefitinib and erlotinib are the oral tyrosine kinase inhibitors (TKIs) (against the intracellular portion of the epidermal growth factor receptor) that have been well studied and are primarily used in the treatment of NSCLC (Fig. 58-3). These agents are associated with an uncommon but sometimes very severe syndrome of interstitial pneumonitis. This pneumonitis appears to be more common in the Japanese population, with a reported incidence of about 3%.¹²⁹ Common radiographic appearance in patients with confirmed gefitinib-induced pneumonitis are nonspecific ground-glass opacities, multifocal areas of airspace consolidation, and patchy ground-glass opacities with septal thickening.¹²⁹ In the TRIBUTE trial (paclitaxel and carboplatin with or without concurrent erlotinib) in persons with advanced NSCLC, erlotinib was not associated with an increased risk of serious pulmonary adverse events.¹³⁰ However, a blinded independent review of interstitial lung disease associated with death in this trial identified fatal interstitial lung disease in 0.6% of cases treated with the combination of erlotinib plus chemotherapy.¹³¹

Other Tyrosine Kinase Inhibitors

The risk of pneumonopathy from targeted therapies does not appear to be limited to the lung cancer population. Imatinib, an inhibitor of KIT and BCR-ABL tyrosine kinases that is used for several types of hematologic malignancies and gastrointestinal stromal tumors, has been associated with interstitial pneumonitis as well.¹³² Dasatinib, a second-generation TKI that inhibits BCR-ABL and SRC kinase, is used for newly diagnosed and imatinib-resistant chronic myelogenous leukemia. Exudative pleural effusions developed in up to 35% of patients treated with dasatinib in phase 1/2 studies.¹³³ Dasatinib can cause pulmonary hypertension (PH) by a direct and specific effect on pulmonary vessels; in a review of French PH registry, nine cases of dasatinib-induced PH were identified.¹³⁴ Crizotinib is another TKI that is approved for the treatment of advanced or metastatic NSCLC and contains a characteristic EML4-ALK fusion oncogene. Severe, life-threatening pneumonitis occurred in 4 of 255 patients (1.6%) treated with crizotinib in a phase 2 clinical trial.¹³⁵ All the cases developed within 2 months of the initiation of treatment. Sorafenib and sunitinib are small molecule multikinase inhibitors of vascular endothelial growth factor and platelet derived growth factor. Both are currently approved for treatment of advanced renal cell cancer (RCC). Interstitial pneumonitis has been reported with sorafenib, which completely resolved upon discontinuation of the drug.¹³⁶ Radiation recall pneumonitis was reported in one case of sunitinib 3 weeks after completion of chest radiotherapy. At this time, there are no specific contraindications to the use of signal transduction inhibitory agents based on pulmonary risk; however, current protocols call for the drugs to be discontinued if thoracic imaging reveals evidence of interstitial pneumonitis.

Monoclonal Antibodies

Many newer molecularly targeted cancer drugs are antibodies against growth factor ligands or cell surface receptors. Bevacizumab is the prototype antibody against vascular endothelial growth factor that is currently approved for use in persons with metastatic colorectal cancer, NSCLC, RCC, and glioblastoma multiforme. Bevacizumab has been associated with life-threatening pulmonary hemorrhage,¹³⁷ which seems to be more associated with centrally located squamous cell lung carcinomas and might be related to rapid tumor response, resulting in bronchovascular fistula. Further studies and means of predicting who is at risk for this devastating complication and how to avoid it are necessary to advance the use of this class of agents.

Cetuximab and panitumumab are monoclonal antibodies that target epidermal growth factor receptor and have activity against metastatic colorectal cancer and squamous cell carcinoma of the head and neck. Cetuximab infusion can cause severe hypersensitivity reactions, including bronchospasms in up to 3% of patients.¹³⁸ Both cetuximab and panitumumab have rarely been associated with pulmonary fibrosis.^139,140 Rituximab and ofatumumab are anti-CD20 monoclonal antibodies approved for the treatment of CD20-expressing hematologic malignancies. Case reports of fatal interstitial pneumonitis have been reported with both agents.^141,142 Rituximab infusion can cause bronchospasm similar to that seen with other chimeric antibodies (i.e., cetuximab), which typically resolves upon stopping or slowing the infusion rate. Fatal pulmonary interstitial fibrosis has also been reported after initiation of weekly rituximab for follicular lymphoma.¹⁴³ Alemtuzumab is a humanized anti-CD52 antibody that is approved for refractory chronic lymphocytic leukemia. Infusion-related pulmonary symptoms of dyspnea and hypoxia have been reported in up to 28% of patients.¹⁴⁴ Diffuse alveolar hemorrhage has been reported with alemtuzumab in the postrenal transplant setting.¹⁴⁵

Ipilimumab is a recombinant, human monoclonal antibody that binds to the cytotoxic T-lymphocyte–associated antigen 4 and blocks interaction with its ligands, CD80/CD86. Inhibition of cytotoxic T-lymphocyte–associated antigen 4 augments T-cell activation and mediates antitumor immune responses. Ipilimumab was recently approved for treatment of advanced melanoma. Severe immune-mediated adverse reactions including pneumonitis have been reported.¹⁴⁶ Although pneumonitis responds to a prolonged course of corticosteroids, permanent discontinuation of ipilimumab treatment is recommended.

Inhibitors of Mammalian Target of Rapamycin

Mammalian target of rapamycin (mTOR) is a serine threonine kinase that regulates cell growth, proliferation, and apoptosis. Noninfectious pneumonitis, a class effect of rapamycin derivatives, develops in patients treated with mTOR inhibitors such as everolimus and temsirolimus. Everolimus is approved for the treatment of RCC, neuroendocrine tumors of pancreatic origin, and subependymal giant cell astrocytoma. Temsirolimus is approved for the treatment of RCC. In clinical studies with these agents, some pneumonitis cases have been severe and occasionally fatal. Radiographic findings consistent with drug-induced pneumonitis were detected in 36% of patients receiving temsirolimus for advanced neuroendocrine tumors and endometrial carcinoma.¹⁴⁷ Only one half of the patients with radiographic changes had clinical symptoms of pneumonitis, and drug treatment was continued in some cases without worsening of the pneumonitis.¹⁴⁷ Guidelines have been suggested for management of temsirolimus-induced pneumonitis (see Table 58-7).

In a placebo-controlled randomized trial of everolimus for advanced RCC, clinical pneumonitis was suspected in 37 (13.5%) of the 274 patients who received this drug.¹⁴⁸ Nine patients (3.3%) had grade 1 pneumonitis (asymptomatic), 18 (6.6%) had grade 2 pneumonitis (not interfering with daily living), and 10 (3.6%) had grade 3 pneumonitis (interfering with daily living or oxygen dependent). Half of the patients with grade 3 pneumonitis had baseline radiologic evidence of pneumonitis before starting treatment with everolimus. The common radiographic features of pneumonitis were focal areas of consolidation at the lung bases or diffuse ground-glass opacities. Resolution of pneumonitis occurred after dose reduction for 20 patients and treatment discontinuation for 10 patients.¹⁴⁸

The etiology of the pneumonitis related to mTOR inhibitors remains unclear. Clinicopathological patterns include interstitial pneumonitis with or without fibrosis, bronchiolitis obliterans organizing pneumonia, or alveolar hemorrhage.¹⁴⁸ Immune-mediated mechanisms of pneumonitis are supported by lung biopsies, bronchoalveolar lavage findings, and the observed clinical response to corticosteroids. The bronchoalveolar lavage fluid and lung biopsy specimens obtained from four patients in whom grade 3 interstitial lung disease developed while they were being treated with everolimus for advanced RCC showed lymphocytosis, interstitial lymphocytic infiltration, and septal thickening of alveolar walls.¹⁴⁹

Other Agents: Thalidomide and Lenalidomide

Thalidomide and lenalidomide are used in the treatment of multiple myeloma and share a complex mechanism of action that includes immunomodulatory and antiangiogenic effects. The most common pulmonary adverse effect of thalidomide and lenalidomide is pulmonary embolism. Venous thromboembolic disease occurs in as many as 12% of patients taking these agents and appears to be higher when the drug is combined with dexamethasone or other chemotherapeutic agents.¹⁵⁰ Dyspnea in association with thalidomide treatment has been reported in 4% to 54% of patients.¹⁵¹ Rare adverse effects with these agents include interstitial pneumonitis, organizing pneumonia, and pleural effusion that typically resolve with steroids and discontinuation of the drug.^152,153

Bortezomib

Bortezomib, a proteasome inhibitor, is currently approved as front-line therapy for multiple myeloma. Bortezomib can cause serious lung injury, and its incidence might vary among different ethnicities. In a series of 13 Japanese patients treated with bortezomib for multiple myeloma, severe pulmonary toxicity developed in four patients and two died of respiratory failure without progression of underlying disease.¹⁵⁴ Autopsy results from a patient in one of the cases showed acute and organizing diffuse alveolar damage. Another case report described a rapidly progressive bronchiolitis obliterans organizing pneumonia resulting in respiratory failure after treatment with bortezomib. This patient responded rapidly to high-dose glucocorticoids.¹⁵⁵