Pomalidomide-induced acute eosinophilic pneumonia: a case report Authors: Mathieu Megevet, Jessica Freitas Pereira, Jeanne Vervier, Pierre-Yves Lovey, Vincent Soubeyran, Pierre-Olivier Bridevaux Published: 2026 ABSTRACT AbstractEosinophilic pneumonia secondary to drug exposure is a recognized but infrequent complication associated with a variety of medications, including oncological therapies. Among immunomodulatory drugs, pulmonary toxicity has been described for thalidomide and lenalidomide but remains a rare complication.Pomalidomide is a third-generation immunomodulatory agent used in relapsed or refractory multiple myeloma. It has been authorized in Europe since 2013. To date, only a limited number of cases of pomalidomide-induced lung injury, and even fewer cases of eosinophilic pneumonia, have been documented in the literature.We present a case of eosinophilic pneumonia attributed to pomalidomide in a 67-year-old patient receiving daratumumab, pomalidomide, and dexamethasone for relapsed multiple myeloma. The diagnosis was supported by compatible respiratory symptoms and imaging, bronchoalveolar lavage eosinophilia, exclusion of alternative causes, lack of clinical improvement with empirical antibiotics, and rapid improvement after pomalidomide withdrawal and systemic corticosteroid therapy.Prompt recognition of this rare adverse event, discontinuation of pomalidomide, and corticosteroid therapy resulted in rapid clinical and radiological improvement, with sustained pulmonary function recovery at 12-month follow-up.Keywords: Pomalidomide, Drug-induced lung injury, Acute eosinophilic pneumonia, Pulmonary toxicity, Multiple myeloma Case presentation A 67-year-old male with a history of relapsed multiple myeloma, previously treated with two autologous hematopoietic stem cell transplantations 8 and 5 years before, was receiving daratumumab, pomalidomide, and dexamethasone (DPD). He was referred to the emergency department by his general practitioner after one week of worsening dyspnea, non-productive cough, sweating, and chills. Pomalidomide had been administered at a dose of 4 mg once daily on days 1–14 of a 21-day cycle. First respiratory symptoms had begun three months earlier, four weeks after DPD initiation, corresponding to a cumulative pomalidomide exposure of approximately 88 mg. On admission, vital signs were notable for a blood pressure of 107/65 mmHg, heart rate of 65 beats per minute, respiratory rate of 15 breaths per minute, temperature of 36.5 °C, and oxygen saturation of 93% while receiving 4 L/min supplemental oxygen via nasal cannula. Laboratory investigations revealed a raised C-reactive protein level at 155 mg/L and peripheral eosinophilia with an absolute eosinophil count of 0.8 G/L. Historical eosinophil counts had been within the normal range. A chest radiograph showed ill-defined bibasilar infiltrates, predominantly affecting the left lower lobe, with preservation of the costophrenic angles. Microbiological tests, including a respiratory virus polymerase chain reaction (PCR) panel, urinary antigen tests for Streptococcus pneumoniae and Legionella pneumophila, and blood cultures, were negative. The patient was admitted to the internal medicine department and empirical antibiotic therapy with co-amoxicillin and clarithromycin was initiated. Clarithromycin was discontinued after 48 h in the absence of initial evidence for atypical bacterial infection. Despite five days of antibiotic therapy, no clinical improvement was observed. A computed tomography (CT) scan of the chest (Fig. 1a) demonstrated diffuse bilateral intralobular reticulations with areas of ground-glass opacities. An arterial blood gas analysis revealed hypoxaemia without hypercapnia or acidosis. Chest CT at diagnosis and after treatment a Coronal CT image at diagnosis showing bilateral ground-glass opacities and intralobular reticulations, highlighted by arrows b One-month follow-up CT showing near-complete resolution after pomalidomide withdrawal and systemic corticosteroid therapy Bronchoscopy with bronchoalveolar lavage (BAL) was performed, revealing a total cell count of 805,000 cells/mL, with a differential showing 32% eosinophils, 48% lymphocytes, 14% macrophages, and 5% neutrophils. Autoimmune testing, including anti-nuclear antibodies and anti-neutrophil cytoplasmic antibodies, was negative. Microbiological analysis of the BAL fluid was positive for Mycoplasma pneumoniae by PCR. PCR tests for respiratory viruses and Pneumocystis jirovecii were negative. In light of the clinical presentation, imaging findings, BAL eosinophilia, and exclusion of other likely causes, a diagnosis of pomalidomide-induced eosinophilic pneumonia was suspected. By the time of discontinuation, the patient had completed four full cycles and had received an additional seven days of cycle 5, corresponding to a cumulative pomalidomide dose of 252 mg. Pomalidomide was permanently discontinued and no re-challenge was performed. The patient was started on systemic corticosteroid therapy with prednisone at 0.75 mg/kg daily (60 mg), which was subsequently tapered over eight weeks. Rapid clinical and symptomatic improvement followed corticosteroid initiation. Pulmonary function tests performed at diagnosis (Table 1) showed lower values than the patient’s pre-illness baseline pulmonary function tests obtained one year earlier, with decreases of 830 mL in forced expiratory volume in one second (FEV1), 1.1 L in forced vital capacity (FVC), and 720 mL in total lung capacity (TLC). Table 1Evolution of pulmonary function tests at baseline, diagnosis, one-month follow-up, and 12-month follow-upParameterMetricBaselineDiagnosis1-monthfollow-up12-monthfollow-upFVC (L)Value5.524.424.524.86% predicted124101105114Z-score1.600.080.330.89FEV1 (L)Value4.553.723.713.95% predicted134112113122Z-score2.300.770.851.39FEV1/FVC (%)Value82848281% predicted108110107106Z-score0.821.120.790.67TLC (L)Value7.306.586.957.70% predicted1019298109Z-score0.11-0.67-0.160.68VC (L)Value5.474.344.595.20% predicted1139198112Z-score1.01-0.70-0.170.95DLCO(mmol/min/kPa)Value8.67.06.78.4% predicted96797796Z-score-0.23-1.29-1.43-0.22KCO(mmol/min/kPa/L)Value1.11.11.11.2% predicted78817785Z-score-1.40-1.22-1.49-0.94Predicted values and Z-scores were calculated according to Global Lung Function Initiative (GLI) reference equationsAbbreviations: FVC Forced vital capacity, FEV1 Forced expiratory volume in one second, TLC Total lung capacity, VC Vital capacity, DLCO Diffusing capacity of the lungs for carbon monoxide, KCO Carbon monoxide transfer coefficient Evolution of pulmonary function tests at baseline, diagnosis, one-month follow-up, and 12-month follow-up DLCO (mmol/min/kPa) KCO (mmol/min/kPa/L) Predicted values and Z-scores were calculated according to Global Lung Function Initiative (GLI) reference equations Abbreviations: FVC Forced vital capacity, FEV1 Forced expiratory volume in one second, TLC Total lung capacity, VC Vital capacity, DLCO Diffusing capacity of the lungs for carbon monoxide, KCO Carbon monoxide transfer coefficient At one-month follow-up, the patient reported no respiratory symptoms. Blood eosinophilia had resolved and repeat CT (Fig. 1b) showed near-complete resolution of prior abnormalities. Pulmonary function testing showed lung volumes remaining within the normal range, with a trend toward higher values and preserved gas transfer. Although gas transfer remained within the normal range, DLCO remained below baseline, while KCO was lower by percent predicted and Z-score, possibly reflecting residual alveolar-capillary dysfunction despite near-complete radiological resolution. At 12-month follow-up, after completion of corticosteroid therapy, the patient remained free of respiratory symptoms, with no recurrence of pulmonary manifestations. Pulmonary function testing showed sustained improvement (Table 1). Discussion Pomalidomide is a third-generation immunomodulatory agent used for the treatment of relapsed or refractory multiple myeloma and has also demonstrated activity in symptomatic Kaposi sarcoma [1–3]. Drug-induced interstitial lung disease has heterogeneous mechanisms and clinical patterns, including direct epithelial injury, immune-mediated inflammation, hypersensitivity-like reactions, and eosinophilic pneumonia [4]. Among immunomodulatory drugs, pulmonary toxicity has been reported with thalidomide [5, 6] and lenalidomide [7–9], with reported patterns including hypersensitivity pneumonitis, organizing pneumonia, interstitial pneumonitis, and eosinophilic pneumonia. The precise mechanisms of immunomodulatory drug-induced pulmonary toxicity remain incompletely understood, but BAL lymphocytosis or eosinophilia, steroid responsiveness, and improvement after drug discontinuation support an immune-mediated or hypersensitivity-like process in many cases. Drug-induced eosinophilic pneumonia is usually supported by compatible symptoms and imaging, BAL eosinophilia, exclusion of alternative causes, and improvement after withdrawal of the suspected drug, with or without corticosteroid therapy [10, 11]. In our patient, accepted diagnostic elements were fulfilled: acute worsening of respiratory symptoms on a subacute background, new pulmonary infiltrates with ground-glass opacities and intralobular reticulations on CT, BAL eosinophilia exceeding 25% (32%), exclusion of likely infectious and autoimmune alternative causes, and clinical and radiological improvement after pomalidomide withdrawal and corticosteroid therapy. In comparison with thalidomide and lenalidomide, pulmonary toxicity associated with pomalidomide appears less frequently reported. Published cases include interstitial pneumonitis, diffuse alveolar hemorrhage, lymphocytic or eosinophilic BAL patterns, and broader pulmonary toxicity [12–18]. Symptoms are usually non-specific, including dyspnea, cough, fever, hypoxaemia, and bilateral ground-glass or interstitial changes on CT, thereby frequently mimicking infection. Most reported patients improved after pomalidomide withdrawal, with or without systemic corticosteroids. Our case shares several features with these reports, including delayed respiratory symptom onset after pomalidomide initiation, bilateral CT abnormalities, BAL eosinophilia, lack of clinical improvement under empirical antibiotics, and rapid improvement after pomalidomide withdrawal and corticosteroid therapy. The positive BAL PCR for Mycoplasma pneumoniae represents the main diagnostic confounder. The exact method was PCR performed on BAL fluid. A nasopharyngeal PCR performed five days earlier was negative. Serological confirmation was not performed, because the overall clinicobiological pattern was not considered consistent with active Mycoplasma pneumoniae as the primary driver, and because serology would not have reliably resolved causality in this immunocompromised patient after PCR detection. The isolated BAL PCR result could therefore represent clinically insignificant detection, contamination, or detection unrelated to the dominant pulmonary syndrome. More generally, nucleic acid detection of Mycoplasma pneumoniae in respiratory samples does not by itself establish symptomatic infection or causality [19]. Several arguments support exclusion of Mycoplasma pneumoniae as the primary cause of the pulmonary syndrome: the symptoms developed after pomalidomide initiation and progressed over several weeks; BAL showed substantial eosinophilia rather than a typical bacterial profile; autoimmune testing was negative; no other infectious cause was identified; the patient did not improve clinically during empirical antibiotic therapy; and improvement was rapid after pomalidomide withdrawal and systemic corticosteroid initiation. Although the short exposure to clarithromycin means that antibiotic non-response alone cannot exclude Mycoplasma infection, the full clinical, biological, radiological, and therapeutic pattern strongly supports pomalidomide-induced eosinophilic pneumonia as the most likely diagnosis. A retrospective causality assessment using the Naranjo Adverse Drug Reaction Probability Scale yielded a score of 7, supporting a probable causal relationship between pomalidomide and the pulmonary event [20]. The score was driven by previous reports of similar pulmonary toxicity, temporal association after pomalidomide exposure, improvement after drug discontinuation and corticosteroid therapy, lack of a more likely alternative cause, and objective evidence including CT abnormalities and BAL eosinophilia (Table 2). Table 2Naranjo adverse drug reaction probability scale applied to pomalidomideNaranjo itemAnswerScorePrevious conclusive reports on this reaction?Yes+ 1Did the adverse event appear after the suspected drug was administered?Yes+ 2Did the reaction improve after the drug was discontinued or a specific antagonist was given?Yes+ 1Did the reaction reappear after readministration?Not done0Are there alternative causes that could on their own have caused the reaction?No likely alternative cause+ 2Did the reaction reappear after placebo?Not applicable0Was the drug detected in blood or other fluids in toxic concentrations?Not applicable0Was the reaction more severe when the dose was increased or less severe when the dose was decreased?Unknown0Did the patient have a similar reaction to the same or similar drugs previously?No/unknown0Was the adverse event confirmed by objective evidence?Yes: CT abnormalities and BAL eosinophilia+ 1Total scoreProbable adverse drug reaction7 Naranjo adverse drug reaction probability scale applied to pomalidomide A comparison with previously reported cases of pomalidomide-related lung injury is provided in Table 3. Table 3Previously reported cases of pomalidomide-related eosinophilic pneumonia or related lung injuryAuthor, yearTime to onsetBAL findingsOutcomeGeyer et al. 2011, case 1 [12]480 dNormalRapid improvement; relapse after reintroductionGeyer et al. 2011, case 2 [12]120 d45% eosinophilsRapid improvement; reintroduction without recurrenceSasieta Tello et al. 2013 [13]7 dAlveolar hemorrhageRapid improvement; relapse after reintroductionModi et al. 2015 [14]8 moNRProgressive improvement; relapse after reintroductionGajic et al. 2018 [15]22 mo30% lymphocytesProgressive resolution; persistent fibrotic lesionsIcard et al. 2021 [16]6 d71% lymphocytesRapid resolutionVivien et al. 2023 [17]2 mo25% eosinophilsRapid resolution; no recurrence during follow-upCarter et al. 2024 [18]NR25% eosinophilsClinical improvement reportedPresent case4 wk32% eosinophils, 48% lymphocytesRapid clinical improvement; near-complete radiological resolution at 1 mo; sustained functional recovery at 12 moAbbreviations: BAL Bronchoalveolar lavage, NR Not reported, d days, wk weeks, mo months Previously reported cases of pomalidomide-related eosinophilic pneumonia or related lung injury Abbreviations: BAL Bronchoalveolar lavage, NR Not reported, d days, wk weeks, mo months Prompt recognition of drug-induced pulmonary toxicity is crucial. Continuation of the offending agent in the context of ongoing injury may lead to persistent or irreversible pulmonary impairment, including fibrotic sequelae [15]. Withdrawal of pomalidomide and early corticosteroid therapy in our patient led to rapid clinical recovery, near-complete radiological resolution at one month, and near-baseline gas transfer at 12 months. Notably, daratumumab was maintained after pomalidomide withdrawal without recurrence of pulmonary symptoms. Daratumumab-induced interstitial pneumonitis has been reported [21], but, to our knowledge, no published case of daratumumab-induced eosinophilic pneumonia has been identified. In the present case, the absence of recurrence despite continuation of daratumumab argues against daratumumab as the primary culprit. Conclusion This case highlights a rare but important adverse event associated with pomalidomide use. Clinicians should maintain a high index of suspicion for drug-induced eosinophilic pneumonia in patients presenting with new respiratory symptoms during therapy. Early diagnosis, discontinuation of the causative agent, and corticosteroid therapy when clinically indicated are key to achieving favourable outcomes. Acknowledgements We want to thank all the medical staff for their help in this case. Abbreviations Bronchoalveolar lavage Computed tomography Diffusing capacity of the lungs for carbon monoxide Daratumumab, pomalidomide, and dexamethasone Forced expiratory volume in one second Forced vital capacity Carbon monoxide transfer coefficient Polymerase chain reaction Total lung capacity Authors’ contributions MM contributed to the interpretation of the data and drafting of the manuscript. PB, JF, JV, VS and PL contributed to the interpretation of the data and revised the manuscript. All authors read and approved the final manuscript. Funding Open access funding provided by University of Geneva. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate Under Swiss law, formal ethics committee approval is not required for publication of a single anonymized case report. Consent for publication Written informed consent for publication was obtained from the patient. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement No datasets were generated or analysed during the current study.