By searching the FBPDB from 2018 to 2022, we identified 10 cases. Of these, three were excluded due to incomplete data. Of the seven remaining cases, five were diagnosed with O-ILD (Table 1). Two cases were reclassified as infectious pneumonitis. Radiological data were fully available for three cases.
Table 1
Patients’ clinical and nonclinical characteristics
| Index case (72, Female) | Case 2 (56, Female) | Case 3 (71, Female) | Case 4 (72, Female) | Case 5 (38, Female) |
Tobacco history | None | None | None | None | None |
Past medical history | Asthma | HT | None | Cured breast cancer | None |
Cancer histology | Ovarian high-grade papillary serous carcinoma | Ovarian high-grade papillary serous carcinoma | Ovarian high-grade papillary serous carcinoma | Ovarian high-grade papillary serous carcinoma | Breast high-risk invasive lobular carcinoma |
Stage at diagnosis | IV | IV | III | IV | IIB |
Mutation | Somatic BRCA1 | Germinal BRCA1 | HRD DNA signature | Somatic BRCA1 | Germinal BRCA1 |
Treatment lines | Debulking surgery | Debulking surgery | Carboplatin-Paclitaxel | Debulking surgery | Epirubicin-Cyclophosphamide |
Carboplatin-Paclitaxel | Carboplatin-Paclitaxel + Bevacizumab | Debulking surgery | Carboplatin-Paclitaxel | Paclitaxel |
Bevacizumab + placebo (PAOLA trial) | Bevacizumab | Carboplatin-Gemcitabine-Bevacizumab | Carboplatin-Paclitaxel + Bevacizumab | Mastectomy and axillary clearance + radiotherapy |
Adjuvant surgery | Carboplatin-Pegylated liposomal doxorubicin | Liposomal doxorubicin-Trabectidin | Olaparib 300mg twice a day | Capecitabine |
Olaparib 300mg twice a day | Olaparib 400mg twice a day | Carboplatin-Gemcitabine | | Olaparib 300mg twice a day |
| | | Olaparib 300mg twice a day | | |
Olaparib-induced interstitial lung disease |
Delay of occurrence | 12 weeks | 12 weeks | 6 weeks | 12 weeks | 33 weeks |
Radiological pattern | Gound glass opacities Hypersensitive pneumonitis | Ground glass opacities Hypersensitive pneumonitis | Ground glass opacities Hypersensitive pneumonitis | Ground glass opacities Hypersensitive pneumonitis | Ground glass opacities Hypersensitive pneumonitis |
Bronchoalveolar lavage | 1,000,000 cells/mL (Ly 30%; PNN 23%; PNE 3,9%; Ma 40%) CD4/CD8 = 3.9 | 650,000 cells/mL (Ly 41%; PNN 1%; PNE 2%; Ma 56%) CD4/CD8 = 2 | 260,000 cells/mL (Ly 45%; PNN 48%; PNE 0%; Ma 7%) CD4/CD8 = 1 | NA | 500,000 cells/mL (Ly 60%; PNN 15%; PNE 2%; Ma 18%) |
Autoimmunity | Negative | Negative | Negative | Negative | Negative |
Bacteriology | Urine Ag L/P negative LBA, bronchial aspirate, sputum, blood, and urine cultures sterile | Urine Ag L/P negative LBA, bronchial aspirate, sputum, blood, and urine cultures sterile | Urine Ag L/P negative LBA, bronchial aspirate, sputum, blood, and urine cultures sterile | Urine Ag L/P negative No culture performed | Urine Ag L/P negative LBA, bronchial aspirate, sputum, blood, and urine cultures sterile |
Virology | Negative PCR | Negative PCR | Negative PCR | NA | Negative PCR |
Pneumocystis | Negative DFME and PCR | Negative DFME and PCR | NA | NA | Negative DFME and PCR |
Cardiac examinations | Normal BNP | Normal BNP and TTE | Negative BNP | Negative BNP | Negative BNP |
Antibiotics | Amoxicillin-clavulanic acid (7 days) followed by Piperacillin-Tazobactam (7 days) | Amoxicillin (7 days) Amoxicillin-clavulanic acid (7 days) | Piperacillin-Tazobactam + Clarithromycin (5 days) | No antibiotics | Amoxicillin-clavulanic acid (7 days) |
Steroids | Steroids 120mg, 12 weeks tapering | Steroids 80mg, 6 weeks tapering | Steroids boluses during 72h (625mg) followed by steroids 64mg, 8 weeks tapering | No steroids | Steroids 100mg, 5 weeks tapering |
Olaparib treatment | Stopped | Stopped | Stopped | Stopped | Stopped |
Clinical outcomes | Complete ILD resolution | Complete ILD resolution | Partial ILD resolution | Complete ILD resolution | Complete ILD resolution |
Rechallenge PARPi | Relayed with Rucaparib | No | No | Rechallenged with olaparib | NA |
Ag L/P: antigens legionella/pneumoccocus; BNP: brain natriuretic peptide ; CD4/CD8: ratio of T helper cells/cytotoxic T cells ; DFME: direct fluorescent microscopy examination ; HT: hypertension; HRD: Homologous recombination deficiency; ILD: interstitial lung disease; BAL: bronchoalveolar lavage ; Ly: lymphocyte ; Ma: macrophage ; NA: not applicable; PNN: polynuclear neutrophil; PARPi: poly(ADP-ribose) polymerase inhibitors; PNE: polynuclear eosinophil PCR: polymerase chain reaction; TTE: transthoracic echocardiogram |
Description of the Index Case
A 72-year woman carrying a somatic BRCA1 mutation diagnosed in 2015 with a metastatic ovarian carcinoma. She never smoked. Debulking surgery was performed, and the patient received first-line platinum-based chemotherapies (Table 1). She was included in the PAOLA trial and randomised to the bevacizumab + placebo arm. Complete radiological response was obtained, and she underwent adjuvant surgery followed by a maintenance treatment with olaparib (300mg twice a day).
Fever, cough, and dyspnoea began twelve weeks after olaparib initiation. CT-scan was initially considered as normal, and olaparib was continued. Retrospective analysis showed a diffuse slight high density of the pulmonary parenchyma with a relative respect of lower lobes. The patient was hospitalised three weeks later. Clinical examination discovered bilateral fine inspiratory crackles. She received nasal oxygen therapy at 4L/min maximum. Thoracic CT excluded pulmonary embolism but showed a worsening of lung parenchyma abnormalities with GGOs associated to areas of lobular hypoattenuation and small nodular lesions (Fig. 1). Radiological presentation was typical of HP. C-reactive protein (CRP) was moderately increased (86mg/dl for normal range < 5mg/dL) and procalcitonin was negative. No other biological anomalies were reported. Urine antigens for legionella and pneumococcus were negatives. BAL showed alveolitis (107 cells/ml) with increased lymphoid population (30%) with an over-representation of T helpers (CD4/CD8 = 3.9) and few eosinophils (3.9%) (Table 1). BAL cytopathology did not reveal tumour cells. Bacteriological cultures remained sterile. No pneumocystis cyst was observed, and polymerase chain reaction (PCR) for pneumocystis was negative, as well as Β-D-glucan. PCR for viruses (respiratory virus including Covid-19) and respiratory germs (including mycoplasma) were negative. Autoimmune serologies were negative. Brain natriuretic peptide was normal. No respiratory function exploration was carried out.
High-dose glucocorticoids (2mg/kg prednisone) were started, associated with antibiotics pending the results of microbiological investigations. Oxygen weaning was quickly obtained. Glucocorticoids were initially introduced for a period of 6 weeks with progressive tapering. After 4 weeks of tapering (until the posology of 20mg/day), dyspnoea and fever reappeared. Thus, glucocorticoids posology was increased to 120mg, and treatment was continued with progressive tapering for 8 extra weeks. The CT-scan performed after completing corticosteroid treatment showed a complete resolution (Fig. 1).
Olaparib was not reintroduced. Based on a recent study suggesting an inferior over-risk of PARPi-induced pneumonitis with rucaparib [6], and given the complete response observed with PARPi maintenance, the physician rechallenged with this new PARPi without pneumonitis recurrence. Nevertheless, the cancer relapsed after 9 months of rucaparib; the patient underwent a new platinum-based chemotherapy.
Description and comparison with other cases
The median age of our five patients was 71 years old (range: 38–72) at the AE time (Table 1). None of them were smokers. Four had ovarian carcinoma and one breast lobular carcinoma. Two patients exhibited a germinal BRCA1 mutation, two a somatic BRCA1 mutation, and one a homologous recombination deficiency (HRD). They received three to four chemotherapy lines before olaparib treatment being started. Initial posology ranged from 300mg to 400mg twice a day. The median delay between olaparib introduction and symptom occurrence was 12 (range: 6–33) weeks. Main symptoms were dyspnoea (n = 5), fever (n = 4), and cough (n = 3). Pneumonia severity (CTCAE V5) was Grade 3 in four patients and Grade 2 in one patient.
GGOs found on lung CT-scan turned out to be the most predominant lesions (Figs. 1–3), with a relative respect of lower lobes in two patients (Index case (Fig. 1) and case #3 (Fig. 3)). In Index case, small nodular lesions were present (Fig. 1). We observed symmetrical topography of lesions in all cases. HP was the common pattern.
BAL was performed, revealing lymphocytic alveolitis in all patients. Median CD4/CD8 ratio was 2 (range: 1-3.9). Median CRP concentration was 63mg/dl (range: 43–86). For each case, bacteriology and autoimmunity were negative and the O-ILD diagnosis was established.
Olaparib was stopped for each patient. Oxygen supply was administered in four cases, with a median oxygen flow rate of 4L/min (range: 2–6). Four patients initially underwent antibiotic treatment for 5 to 7 days associated with glucocorticoids. Glucocorticoid posology ranged from 1mg/kg to 2mg/kg with a median tapering period of 8 (range: 5–12) weeks. Boluses were administrated in one patient during 72 hours. No fatal ILD occurred, while ILD resolved completely for four patients and partially for one, with rapid oxygen weaning and home discharge.
One patient was reintroduced to another PARPi (rucaparib), while another was re-administered olaparib a month after discontinuing glucocorticoids 300mg twice daily (initial prescribed dosage). Three weeks after rechallenge, the patient reported progressive Grade 2 dyspnoea and asthenia assumed by the physician to be a relapse. However, no CT-scan was performed. Olaparib was reduced to 250mg twice a day during 15 days, without effect, and then reduced to 200mg twice a day, with an improvement of symptoms. The cancer relapsed four months after rechallenge.
International WHO and FDA databases cases extraction
Of the 40 ILD cases registered after olaparib treatment, 39 occurred in women, all cases were serious. In 90% of cases, olaparib was the only suspected drug. The median age was 69 (range: 34–87) years. The median time between the beginning of olaparib and the AE was 99 (range: 6-815, n = 21) days. Treatment was stopped in most cases (91%). Twenty-nine cases were considered as “recovering” or “recovered” or “recovered with sequelae”; one case was fatal; three cases were “not recovered”, and seven outcomes were “unknown”. There were no available data on cytopathological lung features or radiological patterns.