Bleomycin lung: a case report

Abstract

A 69-year-old gentleman with non-Hodgkin’s lymphoma (stage I), with baseline fibrotic lung changes on CT, received six cycles of R-PMitCebo chemotherapy containing bleomycin. Three months later he presented to the Accident and Emergency Department with progressive dyspnoea, dry cough, pyrexia and generalised lethargy. Chest radiographs showed bilateral lower zone opacities. Clinically, all signs initially pointed to community-acquired penumonia, but he failed to respond to standard treatment for this. Repeat high-resolution CT (HRCT) subsequently showed widespread peripheral interstitial changes consistent with marked fibrotic lung changes. It became apparent that this was in fact bleomycin-induced pulmonary toxicity. The patient rapidly deteriorated and developed type I respiratory failure. Despite intensive steroid treatment, the patient progressively got worse and died in the Intensive Therapy Unit 10 days after admission. Death was directly attributed to pulmonary fibrosis secondary to bleomycin treatment.

BACKGROUND

This case highlights the importance of recognising the potentially fatal toxic side effects of bleomycin, which is a commonly used and very effective anticancer agent employed for several types of tumours. Although bleomycin pulmonary toxicity is a rare complication, it carries a crude mortality rate of up to 3% and this must be considered when deciding what chemotherapy regime a patient should receive. In patients who have risk factors that make them particularly susceptible to its toxic effects, such as underlying lung disease, alternative regimes that exclude bleomycin may be more appropriate.

CASE PRESENTATION

The following case is one that I encountered as a house officer on my Oncology rotation in Brighton.

The patient was a 69-year-old gentleman who was diagnosed with diffuse large B cell non-Hodgkin’s lymphoma in July 2007 (stage I). His other past medical history included angina and peripheral vascular disease. He was an ex-smoker of 32 pack-years and his previous lifelong occupation had been a number plate manufacturer, which involved polishing and grinding metal. There was no known asbestos exposure. Prior to his admission to hospital he had been living with his wife and had been fully independent, although his exercise tolerance had progressively declined to 50 yards over the past few months.

He presented to the Accident and Emergency Department (A&E) in early December 2007 with a 3-month history of progressively worsening dyspnoea, dry cough and generalised weakness. He denied any other symptoms such as chest pain or haemoptysis. Of note, he started his chemotherapy in August 2007 and had completed his final and sixth cycle of chemotherapy 10 days prior to his admission. This consisted of the R-PmitCebo regimen (rituximab–monoclonal antibody to CD20, prednisolone, mitozantone, cyclophosphamide, etoposide, bleomycin and vincristine). The onset of his dyspnoea began after his second course of chemotherapy (a month into treatment). After this the bleomycin was omitted from the regime and he was started on prednisolone 50 mg.

Interestingly, a baseline CT scan of the chest done prior to starting chemotherapy highlighted early peripheral reticular changes in the right lung, with less marked changes at the left lung base and sparing the remainder of the lung. This was thought to be due to underlying fibrotic lung disease, probably related to occupational exposure (see figs 3–5). His actual volume of lymphoma was small, with a single borderline pretracheal node measuring 11 mm in short axis diameter with no other lymphadenopathy seen.

Figure 3.

CT scan of the chest: 15 August 2007—prior to starting chemotherapy. Cross-sections are shown from the upper (fig 3), middle (fig 4) and lower (fig 5) zones of the lungs. From this CT scan we can see the early signs of underlying fibrotic lung disease, including subpleural interstitial thickening and ground-glass opacification.

Figure 5.

Lower zones.

Figure 4.

Middle zones.

On examination in A&E, he was tachypnoeic (respiratory rate (RR) 22), with oxygen saturation (sats) of 94% on air, and pyrexial (38.1°C). On auscultation of the lungs, crackles were heard bibasally, worse on the right.

INVESTIGATIONS

Blood results showed a raised C-reactive protein (CRP; 21 mg/l) but normal white cell count (WCC) of 8.5×10⁹/l and neutrophils of 6.8×10⁹/l. His chest x ray was reviewed and showed right mid and lower zone opacities as well as left lower zone consolidation (fig 2). There had been an increase in the left lower opacity since the previous film (fig 1). These were initially interpreted as changes associated with infective consolidation, given his pyrexia and raised CRP.

Figure 2.

Chest radiographs taken when the patient presented (2 December 2007) with symptoms.

Figure 1.

Chest radiographs taken mid-treatment (26 October 2007).

The most recent HRCT was subsequently reviewed and showed widespread peripheral interstitial shadowing bilaterally (see figs 6–8) much worse than his previous baseline CT (15 August 2007). It was now felt his presentation was likely to be a combination of exacerbation of his lung fibrosis due to bleomycin plus superimposed bronchopneumonia. He was therefore treated with standard intravenous antibiotics for community-acquired pneumonia and maintained on prednisolone.

Figure 6.

High-resolution CT (HRCT) 15 November 2007. This later CT on completion of chemotherapy shows an increase in the amount of interstitial thickening in the upper (fig 6), middle (fig 7) and lower (fig 8) zones, and honey-combing in the lung bases, worse on the right.

Figure 8.

Lower zones.

Figure 7.

Middle zones.

TREATMENT

The day following his admission the on-call house officer was asked to review him on the ward as he had become more unwell. He was more breathless and dropped his sats from 98% to 77% on 35% oxygen. He was tachycardic (110 beats/min), tachypnoeic (RR 32) and had spiking temperatures of 38.5°C. Bibasal crackles were heard, now left more than right. Aterial blood gases (ABG) on 35% oxygen showed type I respiratory failure with pH 7.51, partial pressure of oxygen (PO₂) 4.5 kPa, partial pressure of carbon dioxide (PCO₂) 4.0 kPA, base excess (BE) 1.7, HCO₃⁻ (bicarbonate 23.9 mmol/l). He was put on 15 litres of oxygen, which improved his pH and hypoxia. His chest x ray showed worsening left-sided consolidation with new right-sided changes. CT pulmonary angiogram was negative for pulmonary embolism. Despite resuscitation he became progressively more breathless and hypoxic (sats 78% on 15 litres of oxygen). He was then initiated on more broad-spectrum antimicrobial treatment (septrin, tazocin and clartihromycin) and transferred to the Intensive Therapy Unit (ITU) for ventilatory support.

OUTCOME AND FOLLOW-UP

In the critical care setting he was initiated on continuous pulmonary airway pressure (CPAP) and methylprednisolone, and although the ventilatory support improved his symptoms of respiratory distress, he immediately desaturated off it and became severely dyspnoeic. His case was discussed at length between the ITU and oncology team. It was important to determine how much of his lung disease was irreversible and secondary to bleomycin and how much of it was in fact due to infection and therefore potentially reversible. The Oncology team felt it was mostly due to the bleomycin and it was about the right time after treatment for this to show up (3 months in his case). Also the fact that he was not improving with antibiotics pointed towards established fibrotic lung disease. In view of this irreversibility and in the light of his comorbidities (pre-existing lung disease) and poor left ventricular function (ejection fraction 42%) it was decided that it would not be in his best interests to intubate him or provide CPR (cardiopulmonary resuscitation) if he were to deteriorate. This was despite the good prognosis of his lymphoma. The family accepted this and felt that alleviation of his pain and distress should be the main priority. Specialist input of the palliative care team was requested to obtain optimum symptom control. CPAP was continued but sadly he died after 10 days in ITU. The cause of death was directly attributed to hypoxia secondary to lung fibrosis.

DISCUSSION

Bleomycin is an antibiotic agent with chemotherapeutic properties, originally isolated from the fungus Streptomyces verticillus in 1966¹ and used clinically since the early 1970s. It is used in the treatment of several cancers such as Hodgkin’s and non-Hodgkin’s lymphoma, germ cell tumours, Kaposi sarcoma, cervical cancer and squamous cell carcinomas of the head and neck.^1,2 It exerts its effects by inducing tumour cell death, and also by possibly inhibiting angiogenesis. Its cytotoxic properties result from induction of free radicals by formation of a complex with ferrous ions and molecular oxygen.^1,3–5 These free radicals then cause breaks in DNA, which eventually lead to cell death.

The drug is usually administered intravenously as it is poorly absorbed by mouth. It is eliminated by the kidneys (50–70% excreted unchanged)⁶ and can also be deactivated by the enzyme bleomycin hydrolase, which is found in both normal and tumour cells.⁴ Its half-life is 2–5 h in patients with normal renal function, but this can increase to 30 h in patients with renal impairment.⁶

Bleomycin is a desirable component of chemotherapy regimens due to its broad activity and low myelotoxicity and immunosuppression.⁴ Cure rates for testicular cancer, in combination with cisplatin and etoposide, are as high as 90%.⁴ However, the application of bleomycin can result in fatal side effects, the most serious of which is bleomycin pulmonary toxicity (BPT). This starts off as a pneumonitis and progresses to fibrosis. It appears to be facilitated by macrophages and lymphocytes secreting tumour necrosis factor.⁶ Hypersensitivity to bleomycin is also known, but this is rarer and more amenable to treatment.⁶

The incidence of bleomycin-induced pulmonary fibrosis is 6–10%,⁷ and in germ cell malignancies a death rate of 2.8% has been attributed to this.⁸ There appear to be a number of common predisposing factors.

From 1982 to 1999 researchers at the Royal Marsden Hospital in Sutton, UK carried out an important study of all the male testicular cancer patients receiving bleomycin therapy (835 patients) and identified those who were at greatest risk of developing BPT. The risk factors for developing BPT that were derived from this study included older age (age >40 years), cumulative dose of bleomycin (>300 000 IU), low glomerular filtration rate (<80 ml/min) and more advanced disease (stage IV disease at first bleomycin dose).

Of these, perhaps the most well known are cumulative drug dose and renal function. High cumulative doses of bleomycin are sometimes used for disease with poor prognosis or as rescue therapy, but have been shown to increase the chances of developing BPT.⁹ As the drug is excreted by the kidney, impaired renal function obviously enhances the half-life of the drug, leading to greater pulmonary exposure. According to the authors, more widespread cancer may be associated with greater total drug dose and intensity, as well as para-aortic metastases that may compromise renal function. It may also indicate the existence of lung metastases, which can increase toxicity to the drug.⁶

In addition to the above, high inspired oxygen concentrations can also precipitate BPT, even several years following treatment with bleomycin.^10,11 It is unknown what the threshold of oxygen exposure is for this or if a time period exists after treatment beyond which oxygen exposure has no effect. It still also remains uncertain whether higher perioperative oxygen concentrations during surgery are a risk factor for toxicity following prior bleomycin treatment.¹² However, it is generally accepted that keeping oxygen exposure to a minimum for these patients is in their best interests. Other suggested risk factors for BPT include cigarette smoking and concomitant radiotherapy of the chest.¹

Symptoms of BPT are often non-specific and usually manifest between 1 and 6 months following bleomycin treatment, with 20% occurring later.¹³ Features include non-productive cough, breathlessness, pleuritic chest pain, fever and tachypnoea.¹ The differential diagnosis of BPT includes pneumonia, other opportunistic lung infections, radiation-induced pulmonary fibrosis, metastatic disease and adverse reaction to other drugs. Lung function tests show a restrictive pattern with low diffusion capacity. However, these are not specific for BPT, and routine pulmonary function tests have been shown to be potentially misleading and ineffective in the management of these patients.¹⁴ The diagnosis of BPT is often one of exclusion by sputum analysis, serology or antigen detection for microorganisms. Patients with BPT are often treated for infection unsuccessfully before the diagnosis is confirmed.

Histology from transbronchial biopsy usually reveals a patchy distribution of diffuse alveolar damage, with an inflammatory element. In terms of imaging, chest radiographs can have variable appearances. The classic pattern is one of bilateral bibasal infiltration with reticulonodular or alveoli-interstitial densities. This can lead to progressive consolidation and honey-combing. HRCT is more sensitive and defines the distribution of the lesions better.¹⁵ Nonetheless, this can also show a spectrum of appearances, depending upon the type of reaction induced by the bleomycin and the underlying histological pattern. The possibilities are broad and include diffuse alveolar damage (DAD), where widespread bilateral ground-glass opacities are seen and dependent areas of airspace consolidation.¹⁶ Similar radiological appearances are noted in acute respiratory distress syndrome. Also reported are hypersensitivity responses, a bronchiolitis obliterans organising pneumonia (BOOP)-type reaction and a chronic pneumonitis picture with lung fibrosis.¹⁶ These all produce variable patterns of airspace, nodules, ground-glass and reticulation. Therefore, it is essential to interpret the HRCT in the full clinical context of the patient as radiological findings may not be distinctive on their own.

There is currently no standard accepted treatment for BPT, although the most common approach involves stopping bleomycin if this is suspected. High-dose corticosteroids are then the mainstay of treatment and the dose should depend on the severity of the pneumonitis^1,17 (eg, prednisolone 1 mg/kg tapering down over a period of months). Following this, patients should continue on a bleomycin-free regimen. When patients survive the initial episode, they almost always recover completely with normalisation of pulmonary function and disappearance of signs and symptoms. The overall mortality rate is 1–3%.^1,6,8,9

In Hodgkin’s lymphoma, BPT has been shown to lower 5-year overall survival rates significantly.¹⁸ However, Martin et al also noted that in patients with Hodgkin’s lymphoma who did not die from acute lung injury due to bleomycin, proceeding without bleomycin had no impact on overall survival and progression-free survival.¹⁸

The interesting thing about this case was that the patient actually had low volume stage I disease (Ann Arbor Classification) from the point of view of his lymphoma and was therefore curable in this respect. He also had low total exposure to the drug and his condition failed to respond to steroids and withdrawing the bleomycin treatment. In total the patient received 60 000 units of bleomycin. As discussed above, bleomycin-related lung fibrosis is usually due to cumulative doses of >300 000 units.⁶ The interstitial disease may be worsened by a high FiO₂ (fractional inspired oxygen), but again this is usually related to higher doses of bleomycin. Although cases of enhanced pulmonary toxicity have been described at low cumulative doses of bleomycin (<300 000 units), these have usually been reversible, and deaths are rare at doses <100 000 units.¹⁹ However, they are not unheard of, and McLeod et al described the case of a patient with non-Hodgkin’s lymphoma who died as a result of receiving a cumulative dose of 60 000 units of bleomycin.¹⁹ However, the patient did have end-stage renal failure and it was thought his concomitant cyclophosphamide treatment may have had a synergistic effect on lung toxicity.

Factors that pointed towards the diagnosis of BPT in our patient:

1. Known exposure to the drug.

2. His risk factors including pre-existing lung disease as seen on CT prior to starting treatment, probably occupationally related. Also his older age probably played a role.

3. Clinical course and time frame of symptoms in relation to him starting his chemotherapy.

4. Non-specificity of symptoms and failure to respond to antimicrobial treatment for pneumonia.

5. Classical pattern of infiltrates in the lower lobes of the lungs.

Possible alternative diagnoses:

1. Pneumonia. This could have been superimposed and included opportunistic fungal infection for instance, which is important to consider in patients receiving chemotherapy who may have an impaired immune system. However, the patient was not immunocompromised at the time and did not respond to any treatments for this.

2. BOOP. In this condition, chest radiographs also show patchy bilateral alveolar infiltrates, whilst HRCT shows mainly peripheral and peribronchial consolidation. It is, however, difficult to diagnose without a lung biopsy.

3. Bleomycin-induced hypersensitivity pneumonitis, but both this and BOOP have a good response to steroids.¹ Also hypersensitivity usually occurs within hours of initiation of treatment and is a more acute reaction. A peripheral and/or pulmonary eosinophilia may also be seen.

4. Cryptogenic fibrosing alveolitis, given the radiological appearances, but again unlikely given the history and clinical course.

5. Cyclophosphamide-induced lung injury. Concomitant cyclophosphamide may potentiate lung injury in patients receiving bleomycin. However, the precipitant of toxicity is almost always the bleomycin, and reducing the dose of this alone has been shown to reduce pulmonary toxicity in patients.¹⁹

To conclude, bleomycin is a very effective antineoplastic agent in non-Hodgkin lymphoma as well as various other malignancies. However, its dose-limiting side effects such as acute lung toxicity are widely feared. Treatment with bleomycin may adversely affect the outcome of cancer patients, in particular those with notable risk factors, who otherwise have a good prognosis. Lung complications can be fatal, as demonstrated in the case above. Extreme caution should be exercised in this group, for instance those with underlying lung disease. If patients have significant risk factors or develop signs of BPT during treatment, alternative chemotherapy regimens should be initiated that omit bleomycin, and all patients should begin a course of steroids. Recognising the features of BPT at an early stage is vital, and in most patients who survive recovery is made without long-term pulmonary sequelae. However, a crude mortality of up to 3% is clinically significant and patients should be fully informed and educated about the risks before embarking on chemotherapy.

LEARNING POINTS

• To appreciate the toxic side effects of commonly used cytotoxic chemotherapy agents such as bleomycin.

• To evaluate a patient fully for predisposing risk factors such as underlying lung disease or renal impairment before embarking on chemotherapy containing bleomycin. If significant risk factors are present an alternative regimen should be commenced.

• To recognise the signs of pulmonary toxicity early on in treatment so that prompt withdrawal of the culprit agent can take place and steroid treatment be initiated.