Overcoming T790M mutant small cell lung cancer with the third-generation EGFR-TKI osimertinib.

A large number of EGFR mutant non-small cell lung cancer patients primordially benefit from first-line treatment with first-generation EGFR-tyrosine kinase inhibitors (TKIs), such as gefitinib and erlotinib. However, multiple acquired resistance mechanisms have been described that limit the clinical efficacy of first-generation EGFR-TKIs. Herein, we report a rare case of lung adenocarcinoma harboring an EGFR exon 19-deletion mutation before the administration of target therapy. This patient acquired resistance to first-generation EGFR-TKIs through small cell lung cancer (SCLC) transformation accompanied by the T790M mutation. Unexpectedly, this SCLC patient maintained a sensitive response to the third-generation EGFR-TKI osimertinib. This special case may indicate that osimertinib represents an effective target drug for SCLC patients who harbor an EGFR T790M mutation.

Introduction

EGFR‐TKIs have revolutionized the therapeutic field by inhibiting the EGFR‐induced downstream signaling pathway in non‐small cell lung cancer (NSCLC). However, most patients ultimately experience drug resistance and progression within two years, and several acquired resistance mechanisms have been identified, including EGFR T790M mutation, MET or HER2 amplification, phosphoinositide 3‐kinase pathway activation, and rare transformation from lung adenocarcinoma (LUAD) to small cell lung cancer (SCLC).1, 2, 3, 4 If an NSCLC patient has acquired a T790M mutation, the third‐generation EGFR‐tyrosine kinase inhibitor (TKI) osimertinib is recommended.5

Case report

A 46‐year‐old woman with a passive smoking history who presented with dyspnea and unintentional weight loss underwent chest computed tomography (CT) scanning that revealed a right middle lobe mass (51 x 61 mm) and multiple thorax and bone metastases. Her clinical stage was T4N2M1b (stage IV). Morphologically, immunohistochemistry showed a poorly differentiated adenocarcinoma that was diffusely positive for thyroid transcription factor 1 (TTF‐1) and focally positive for Napsin A (Fig 1a–c). The tumor harbored a classic EGFR exon 19‐deletion mutation, as shown by qualitative detection (amplification refractory mutation system PCR) (Fig 1a–d). The woman was subsequently treated with gefitinib and experienced significant regression of the mass.

Figure 1

First biopsy specimen: (a) hematoxylin and eosin staining was diffusely positive for (b) thyroid transcription factor 1, (c) focally positive for Napsin A, and (d) harbored a classic EGFR exon 19‐deletion mutation.

Six months after her first exposure to gefitinib, the primary mass and metastatic nodules exhibited obvious enlargement, accompanied by gradually increasing bone pain. The patient subsequently underwent CT‐guided percutaneous lung biopsy, which revealed SCLC transformation (Fig 2a). Given that sufficient biopsy tissue was not obtained for immunohistochemistry staining and next generation sequencing (NGS), a blood sample was collected for driver gene testing and revealed that she harbored a new EGFR‐T790M mutation (Fig 2b). She was then administered osimertinib therapy and experienced excellent improvement.

Figure 2

(a,b) Second biopsy specimen: hematoxylin and eosin staining exhibited EGFR T790M mutation.

However, approximately five months after initial exposure to osimertinib, CT and whole‐body bone scanning revealed further deterioration. Another CT‐guided percutaneous lung biopsy specimen presented SCLC morphology with neuroendocrine markers, including diffuse positivity for Syn and focal positivity for CgA and CD56 (Fig 3). Additionally, peripheral blood was collected, and we defined circulating tumor cells (CK+/CD45‐/4′,6‐diamidino‐2‐phenylindole [DAPI]+) and white blood cells (CK‐/CD45+/DAPI+) (Fig 4a). RB1 and TP53 inactivation mutations were definitively identified in this SCLC transformation (Fig 4b).6 Additionally, tissue NGS also revealed that the EGFR exon 19 deletion (81.18%) and EGFR T790M mutation (3.10%) were retained, and some new mutation positions were detected, including CTNNB1, FGFR2, HRAS, PIK3CA, and RET mutations. The patient was administered a standard chemotherapy strategy with an intermittent etoposide‐cisplatin regimen (EP) and continued to take osimertinib after each cycle of chemotherapy. After six cycles of chemotherapy, chest CT revealed obvious clinical responses, including shrinkage of the primary lung mass and metastatic nodules. The patient continued to take osimertinib, and achieved progression‐free survival (PFS) of four months. In March 2018, the tumor progressed with a chest mass and multiple brain metastases. After written informed consent was obtained, a fourth biopsy of the progressing mass was performed, revealing Syn‐positive, CgA‐positive, and CD56‐positive SCLC with no evidence of adenocarcinoma histology (Fig 5a–d). Molecular analysis revealed that the T790M mutation was retained, and a new EGFR C797S mutation was identified (Fig 5e). Detailed driver gene information and biopsy, treatment, and image scanning history are presented in Figures 6 and 7.

Figure 3

Third biopsy specimen: (a) hematoxylin and eosin staining was diffusely positive for (b) Syn and focally positive for (c) CgA and (d) CD56.

Figure 4

(a) Circulating tumor cells (CBCs) were positive for CK and 4′,6‐diamidino‐2‐phenylindole (DAPI), white blood cells (WBCs) were positive for CD45 and DAPI and (b) harbored RB1 and TP53 inactivation mutations.

Figure 5

Fourth biopsy specimen: (a) hematoxylin and eosin staining was weakly positive for (b) Syn and (c) CgA, (d) diffusely positive for CD56, and (e) harbored EGFR T790M and EGFR C797S mutations. SCLC, small cell lung cancer.

Figure 6

The course of treatment history and driver gene evolution.

Figure 7

Image scanning history: (a,h) before gefitinib treatment; (b,i) response to gefitinib; (c,j) progression after acquired resistance to gefitinib; (d,k) response to osimertinib; (e,l) progression after transformation into small cell lung cancer; (f,m) response to osimertinib‐etoposide‐cisplatin (EP); and (g,n) progression after acquired resistance to osimertinib.

Discussion

In this case, this patient acquired resistance to first‐generation EGFR‐TKI through T790M mutation accompanied by SCLC transformation. Previous studies have demonstrated that T790M mutation is a major cause of resistance to gefitinib in NSCLC.1 As another mechanism of resistance, SCLC transformation is reported to occur in 4–14% of LUAD cases.7 SCLC transformation is associated with the rapid progression and deterioration of lung cancer. Furthermore, there are no ideal targeted drugs for SCLC patients. Although Okamoto et al. reported that SCLC harboring activating EGFR mutations exhibited a surprising response to gefitinib,8 insufficient evidence is available to show that osimertinib could be a targeted candidate for SCLC. Despite SCLC transformation, this case of SCLC still presented sensitivity to osimertinib. We suggest that the T790M mutation may also underlie responsiveness to osimertinib in SCLC. The fourth biopsy in our patient revealed that SCLC harbored a new C797S mutation. Based on previous reports, C797S mutation contributes to osimertinib resistance in NSCLC.1 Thus, C797S mutation may also play a significant role in osimertinib resistance in SCLC.

According to the classic theory of tumorigenesis, every subset of tumor cells unavoidably present characteristics of inter‐tumor heterogeneity between primary tumors and their metastatic lesions, including histological and genomic heterogeneity. Currently, only one biopsy lesion is recommended for histological diagnosis and genomic analysis.9 In our case, the primary mass, pleural effusion cytology, and metastatic lymph nodes simultaneously presented LUAD characteristics. Therefore these observations basically confirmed that this patient had LUAD before gefitinib therapy. PFS after gefitinib treatment was only six months. T790M mutation and SCLC transformation simultaneously accelerated gefitinib resistance.

After acquiring gefitinib resistance, a lung biopsy specimen presented SCLC morphology and RB1 and TP53 mutations; inactivation of both RB1 and TP53 is an effective predictor of transformation from LUAD to SCLC.6 Therefore, the combined histological characteristics and RB1/TP53 mutations in this case indicated SCLC (after gefitinib resistance) rather than LUAD before osimertinib therapy in the second biopsy. Osimertinib is an ideal target drug for gefitinib resistant‐NSCLC patients who harbor EGFR T790M mutations; the median PFS is 10.1 months.5 However, this SCLC case presented further deterioration at approximately five months rather than 10.1 months. We consider that SCLC is a highly aggressive malignant tumor with a rapid growth index and a routinely poor prognosis. The detailed molecular mechanism is very complicated. Further considerations are necessary to uncover the possible inter‐tumor heterogeneity underlying this patient's treatment process. In particular, previous studies have demonstrated that SCLC transformation from adenocarcinoma retains the original EGFR mutation. We discovered that this SCLC maintained the EGFR 19‐deletion in several subsequent biopsies, which is similar to the original EGFR mutation spectrum in LUAD. Thus, we speculate that the SCLC EGFR 19‐deletion mutation in this case was inherited from the original adenocarcinoma.

In summary, our case highlights four points. First, few studies have reported the same patient undergoing four biopsies within the course of EGFR‐TKI treatment with driver gene information. Second, concomitant SCLC transformation and T790M mutation as the manifestation of acquired resistance after first‐line EGFR‐TKI treatment is relatively rare. Third, a case of an osimertinib‐responsive SCLC patient harboring a T790M mutation has not previously been reported, and this patient maintained a clinical response for six months. Fourth, this is the first case report of an SCLC patient with C797S mutation after acquired resistance to osimertinib. Overall, this case may significantly extend the therapeutic territory of osimertinib.

Disclosure

No authors report any conflict of interest.