Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Thorac Oncol. Author manuscript; available in PMC 2017 March 16.
Published in final edited form as:
PMCID: PMC5353353

Transient asymptomatic pulmonary opacities occurring during osimertinib treatment

Sinead A. Noonan, MD, MSc,1 Peter B. Sachs, MD,2 and D. Ross Camidge, MD, PhD*,1



Osimertinib is an Epidermal Growth Factor Receptor (EGFR) Inhibitor licensed for the treatment of EGFR mutant, T790M positive, non-small cell lung cancer (NSCLC). Previously unreported, common, transient asymptomatic pulmonary opacities (TAPOs) were noted at the University of Colorado in patients during osimertinib therapy.


CT imaging and clinical notes of NSCLC patients treated at the University of Colorado with osimertinib were retrospectively reviewed.


Seven of twenty patients (35%), developed TAPOs while on osimertinib. The radiological patterns seen included ground-glass opacities with/without nodular consolidation. The median time to first lesion development was 8.7 weeks (range: 1.6 – 43 weeks) and 6 weeks (range: 1 – 11 weeks) to resolution during continued osimertinib.


TAPOs may be a previously unrecognized, benign feature associated with osimertinib therapy, which may be mistaken for isolated pulmonary progression or the beginning of more severe pneumonitis. If new onset pulmonary lesions, especially those associated with ground-glass appearances, are asymptomatic, localized and there is no evidence of disease progression elsewhere it may be reasonable to continue treatment with osimertinib and monitor these lesions for resolution.

Keywords: osimertinib, pneumonitis, interstitial lung disease, transient asymptomatic pulmonary opacity, ground-glass opacity


Pneumonitis, a type of interstitial lung disease (ILD) is a rare but potentially fatal consequence of treatment with tyrosine kinase inhibitors (TKIs) targeting the epidermal growth factor receptor (EGFR).1,2 Radiological patterns associated with pneumonitis include focal consolidation, peri-bronchovascular ground-glass opacities predominantly in the lung bases, subpleural reticulation and centrilobular nodules with an upper lobe distribution.3,4 Beyond its relationship to EGFR TKIs, pneumonitis can occur secondary to other causes including infection, hypotension, inhalation injury and post-radiation therapy.5 In addition, radiological patterns overlapping with those of pneumonitis can occur secondary to conditions such as pulmonary adenocarcinoma in-situ, lymphangitic carcinomatosis, pulmonary hemorrhage or pulmonary edema.5

Osimertinib is an irreversible EGFR TKI approved for the treatment of EGFR mutation positive non-small cell lung cancer (NSCLC) resistant to first and second generation EGFR TKIs due to the acquisition of a T790M mutation6. The drug has activity against both the common activating EGFR mutations (L858R and exon 19 deletions) and the T790M resistance mutation, with relative sparing of the wildtype form of the EGFR. Patients were excluded from the AURA osimertinib trials with any prior history of ILD. Six cases of potential osimertinib-associated pneumonitis were noted among 253 treated patients.6 Within the trials, treatment discontinuation was mandated for ILD/pneumonitis of any grade and all 6 patients discontinued drug.

This report documents the novel finding of transient asymptomatic pulmonary opacities (TAPOs), occurring among a significant proportion of patients treatment with osimertinib at the University of Colorado. These lesions represent the on-treatment development of clinically benign areas of localized ground-glass opacity in the lungs, with or without a nodular component, which resolve despite continued dosing with osimertinib and therefore appear to have a different natural history from the pneumonitis previously associated with EGFR TKIs.


Clinical and demographic data on all patients treated with osimertinib at the University of Colorado, either within a clinical trial or post-approval, were collected. An institutional review board-approved protocol permits clinical correlates to be made on all University of Colorado patients in whom molecular analyses have been conducted within the Colorado Molecular Correlates laboratory.


University of Colorado Patient Dataset

20 NSCLC patients were treated with osimertinib (19 at 80mg, 1 at 160mg) at the University of Colorado (Table 1). Osimertinib therapy began in individual patients from January 2014 through March 2016. At the cut-off for data analysis (May 26th 2016) the median follow-up on these patients was 26 weeks (range: 11 – 122 weeks). The median scan frequency on osimertinib for the 20 patients was 11 weeks (range 6–12 weeks). Imaging was not standardized and both CT and PET/CT were used in the follow-up of the patients.

Table 1
Demographic and clinical features of osimertinib treated patients who did and did not develop Transient Asymptomatic Pulmonary Opacities (TAPOs) on osimertinib therapy

All patients had a known EGFR activating mutation, 30% (6/20) with L858R, 60% (12/20) with exon 19 deletion and one patient (5%) with both an exon 18 and exon 20 mutation (G719A and S768I). The precise EGFR mutation was not documented for one patient. Two patients waere treated with osimertinib as first-line therapy. All other patients had been previously treated with another EGFR tyrosine kinase inhibitor before the osimertinib. In addition to their initial activating mutation, 88% (16/18) of the previously treated patients had T790M proven as their mechanism of acquired resistance to the prior EGFR inhibitor. All EGFR mutation data were based on tissue biopsies. Two patients were former smokers and six patients had had prior lung or mediastinal radiation (Table 1). None had a prior history of interstitial lung disease (ILD), but two had pre-existing ground-glass opacities on their chest scans and two had a prior history of aspiration.

Seven of the twenty osimertinib treated patients (35%), including the two patients receiving osimertinib in the first-line setting, developed radiographic pulmonary changes on therapy, which were asymptomatic, localized and disappeared and, in some cases, appeared again during continued therapy with osimertinib (Table 1 and Figures 1 and and2).2). Two patients developed new pulmonary changes that persisted and later had a separate identifiable cause, including radiation pneumonitis and disease progression and these were not included in the retrospectively defined cohort of those manifesting TAPOs on therapy (Figure 3). The median time to first development of lesions which, retrospectively, fitted the TAPO description was 8.7 weeks (range: 1.6 – 43 weeks).

Figure 1
Transient Asymptomatic Pulmonary Opacity (TAPOs) occurring during therapy with Osimertinib
Figure 2
Representative images of TAPOs. A. Baseline scan on Patient F. B. Patient F developed an asymptomatic left upper lobe nodule with a ground glass component (with a pure ground glass opacity (GGO) more medially) after 7 weeks on therapy with osimertinib ...
Figure 3
Representative images of pulmonary opacities developing on therapy with osimertinib that were not TAPOs. A. Baseline scan for patient shown in A–C. B. GGO developed 8 weeks following radiation to site of oligoprogression after 38 weeks on osimertinib ...

The sites of the initial TAPOs (9 lesions) across the seven patients showed some bias towards the lower lobes (right upper lobe (1), left upper lobe (1), right middle lobe (1), right lower lobe (2) and left lower lobe (4)). All patients had known deposits/primary sites of cancer in the lungs, however TAPO lesions did not occur coincident with the exact sites of known cancer deposits. Five of the seven patients only had a single TAPO site at initial presentation. Two of the seven patients had synchronous TAPOs occurring at two sites at first presentation. Four patients had TAPOs occurring on >1 occasion (4 patients experiencing a second occurrence (comprising 7 new lesions in total) and 3 patients experiencing a third occurrence (comprising 4 new lesions in total). Of the eleven recurrent lesions, five recurred within the same lobe as the initial lesion. Although recurrent lesions could recur close to previous sites of lesions in the same lobe, the exact site of the opacity did not appear to be identical in these cases.

None of the seven patients had evidence of disease progression elsewhere and all patients were asymptomatic at the time a TAPO occurred. Six of the seven patients had normal oxygen saturations at rest on room air at the time the lesions were first noted (range: 90 to 98% on room air). One patient was on oxygen at baseline and his oxygen requirements did not change during osimertinib treatment or at the time of development of the TAPO. All patients had elevated serum tumor markers at baseline, which decreased in response to therapy with osimertinib. Six patients had baseline elevations in CEA and one patient had an elevated CA27.29. Across the fourteen temporal occurrences of TAPOs, twelve were not associated with any increases in tumor markers from nadir values. Two patients had an elevation in their CEA at the time of their third consecutive TAPO. For one of these, this marker remained within the normal range and did not increase further. For the other, a PET/CT was ordered and did not show disease progression. These lesions subsequently resolved consistent with the behavioral pattern of the TAPOs in the other patients.

The radiological patterns described by a Board-certified radiologist (PBS) in the seven cases included ground-glass opacities, peribronchial nodules and subpleural nodules. Four lesions consisted of ground-glass opacities which also included a nodular consolidation component within the same lesion.

Median time to resolution or improvement in these lesions after initial detection was 6 weeks (range: 1 – 11 weeks). Initially, some of these were attributed to aspiration pneumonia and two patients were treated with antibiotics, with subsequent resolution of the opacities, however both patients manifested later, additional TAPOs reappearing and resolving without antibiotics (Figure 1).

Review of available scans during preceding non-osimertinib EGFR TKI therapy in all 20 patients revealed evidence of similar appearing/disappearing lesions in only 3 cases (only 1 of whom later developed TAPOs on osimertinib). These non-osimertinib associated lesions appeared in these cases while on erlotinib, rociletinib and tesevatinib.


Here we report, seven of twenty patients (35%), developing transient, asymptomatic pulmonary opacities while on osimertinib therapy. The radiological patterns seen in association with the osimertinib-related TAPOs included ground-glass opacities, peribronchial nodules and subpleural nodules. The ground-glass opacities could also include a nodular consolidation component within the same lesion, as might occur in an organizing pneumonia. The percentage of patients with prior thoracic radiation, prior GGOs or a prior history of aspiration was higher among those with TAPOs than among those without (Table 1). In contrast, the percentage of patients with a prior smoking history was lower. Whether the frequency of these lesions could be influenced by some prior lung insult therefore has to be considered, however their specific etiology remains unknown. Similar lesions could be found during prior EGFR TKI therapy, but at a lower frequency (3/20, 15%) and without any apparent association with the development of TAPOs on subsequent osimertinib therapy. The median time to first development of these lesions on osimertinib was 8.7 weeks and the median time to resolution or improvement in these lesions during continued osimertinib exposure was 6 weeks.

A similar case of localized (right upper lobe) ground-glass opacities appearing after 4.5 months on osimertinib has been described, which resolved, coincident with giving the osimertinib on alternate days together with high dose steroids.7 While this case was labelled as osimertinib-induced interstitial lung disease and treated with corticosteroids, it is possible, given its resolution despite continued dosing with the osimertinib, that this could also have represented a TAPO case.

It would be interesting to know if the AURA trial dataset contains evidence supportive of osimertinib-related TAPOs. Technically, lesions meeting the TAPO definition could be coded as Grade 1 pneumonitis per CTCAE version 4.0. However, it is possible that TAPO-like lesions may either have been not reported as adverse events (the patients being asymptomatic), or reported using terms other than pneumonitis/ILD, such as Grade 1 ground-glass opacities (the term used in the Colorado patients who were treated on study) which would not necessitate withdrawing patients from therapy.

As TAPOs could also be mistaken for pulmonary progression, it would also be interesting to note the proportion of cases in the AURA trials manifested progression with only new pulmonary lesions.

Given the range of differential diagnoses for pulmonary lesions in NSCLC patients on therapy, careful workup of any new lesion is important. Symptomatic/progressive ILD/pneumonitis on osimertinib may require prompt supportive care and drug discontinuation to avoid further worsening. However, TAPOs may be a previously unrecognized, benign feature associated with osimertinib therapy which may be mistaken for isolated pulmonary progression or the beginning of more severe pneumonitis. Their etiology remains unknown. If new onset pulmonary lesions, especially those associated with ground-glass appearances, are asymptomatic, localized and there is no evidence of disease progression elsewhere it may be reasonable to continue treatment with osimertinib and monitor these lesions for resolution. Prospective data collection among a larger group of patients, including those on other 1st–3rd generation EGFR inhibitors would be useful to further characterize these events.


DRC is partially supported by the University of Colorado Lung Cancer SPORE (P50CA058187 PI: Paul A. Bunn, Jr.)


Disclosures: None


1. Hong D, Zhang G, Zhang X, Lian X. Pulmonary Toxicities of Gefitinib in Patients With Advanced Non-Small-Cell Lung Cancer: A Meta-Analysis of Randomized Controlled Trials. Medicine (Baltimore) 2016 Mar;95(9):e3008. [PMC free article] [PubMed]
2. Cohen MH, Williams GA, Sridhara R, Chen G, Pazdur R. FDA drug approval summary: gefitinib (ZD1839) (Iressa) tablets. Oncologist. 2003 Jan;8(4):303–6. [PubMed]
3. Travis WD, Costabel U, Hansell DM, King TE, Lynch Da, Nicholson AG, et al. An official American Thoracic Society/European Respiratory Society statement: Update of the international multidisciplinary classification of the idiopathic interstitial pneumonias. Am J Respir Crit Care Med. 2013;188(6):733–48. [PubMed]
4. Macerelli M, Caramella C, Faivre L, Besse B, Planchard D, Polo V, et al. Does KRAS mutational status predict chemoresistance in advanced non-small cell lung cancer (NSCLC)? Lung Cancer. 2014 Mar;83(3):383–8. [PubMed]
5. Camus P, Kudoh S, Ebina M. Interstitial lung disease associated with drug therapy. Br J Cancer. 2004 Aug;91(Suppl 2):S18–23. [PMC free article] [PubMed]
6. Jänne Pa, Yang JC-H, Kim D-W, Planchard D, Ohe Y, Ramalingam SS, et al. AZD9291 in EGFR Inhibitor-Resistant Non-Small-Cell Lung Cancer. N Engl J Med. 2015;372(18):1689–99. [PubMed]
7. Nie KK, Zou X, Geng CX, Zhang L, Liu SC, Zhang CL, Ji YX. AZD9291-induced Acute Interstitial Lung Disease. Chin Med J. 2016;129:1507–8. [PMC free article] [PubMed]