Lung cancer is the main cause of cancer deaths in both men and women in the United States. The disease has 2 major histological subtypes: Non–small cell lung cancer accounts for 85% of cases, with small cell lung cancer comprising the remaining 15%.<sup>1</sup>
Sandhya Sharma, MD, MBBS
Lung cancer is the main cause of cancer deaths in both men and women in the United States. The disease has 2 major histological subtypes: Nonsmall cell lung cancer (NSCLC) accounts for 85% of cases, with small cell lung cancer (SCLC) comprising the remaining 15%.1NSCLC is further divided into adenocarcinoma, squamous cell carcinoma, and large cell carcinoma. Lung adenocarcinoma (LADC) is the most common subtype, accounting for approximately 40% of all lung cancer diagnoses.1,2Targeted therapies have been approved for various mutations seen in LADC. Of these, the EGFR mutation is most common, observed in about 15% of US cases of LADC.3The EGFR mutation is associated with a better prognosis due to the availability of the tyrosine kinase inhibitors (TKIs) that specifically target the tumor cells. However, acquired resistance to these drugs is almost unavoidable, leading to a progression-free survival ranging from 9.6 to 13.7 months.4Various drug resistance mechanisms have been described in the literature, including but not limited to the EGFR T790M mutation.4,5SCLC transformation is a relatively rare acquired drug resistance mechanism in EGFR-mutant LADC that has been treated with TKI therapy.5Identification of this resistance is important because SCLC-specific treatment with platinum-based therapy and etoposide has shown to provide a clinical benefit.4,5We describe a case of EGFR-mutant LADC that transformed to SCLC 2 years after erlotinib (Tarceva) therapy.
We present a case of a 67-year-old woman with remote history of breast cancer who was diagnosed in 1988, for which she was treated with radical mastectomy, adjuvant chemotherapy, and 5 years of tamoxifen therapy. She was noted to have a left lung lesion on a preoperative scan that was performed before breast reconstruction surgery in 2009. A more extensive work-up, including PET/CT scans, showed fluorodeoxyglucose (FDG)-avid left lung mass, left-sided pleural effusion, right lung lesion, and multiple mediastinal lymph nodes. She underwent video-assisted thoracoscopic surgery and biopsy of the left lung mass, which was consistent with LADC. She was treated with cisplatin, paclitaxel, and bevacizumab, to which she had an excellent response for almost 18 months, until follow-up scans identified progression of disease. Repeat biopsy confirmed LADC and sequencing of the EGFR gene revealed an L858R mutation. In September 2011, she was started on erlotinib. She had an excellent partial response to the therapy but stopped the treatment 2 years later due to personal choice. She had been off erlotinib for 1 year when she presented with cough and shortness of breath. A PET/CT scan conducted at this time showed progression of disease, not surprisingly. She underwent a needle biopsy of the left lung lesion, which was consistent with LADC with the EGFR L858R mutation, identical to her prior pathology. The patient experienced postprocedural pneumothorax that required chest tube placement and hospital admission for 1 week. She was restarted on erlotinib at 100 mg daily, which was her previously tolerated dose, to which she again responded for approximately 1 year until October 2016, when she was noted to have progression of disease with increased left lung mass, mediastinal lymphadenopathy, and retrocrural nodes. She also had worsening pulmonary symptoms and increased cough. Peripheral blood was sent for next-generation sequencing, which revealed a known L858R mutation but was negative for T790M mutation. She continued to progress despite an increase in dose of erlotinib to 150 mg daily. A second liquid biopsy also failed to reveal T790M mutation. Ultimately, she underwent a mediastinal lymph node biopsy, which was consistent with SCLC. The patient underwent an MRI of the brain, which was negative for metastatic disease. The patient was started on carboplatin and etoposide. She completed 6 cycles with postchemotherapy PET/CT scan showing partial response. Despite an excellent response, there were still areas of FDG activity suggestive of active disease. In addition, carcinoembryonic antigen kept trending up. Persistent EGFR L858R mutation with no secondary resistance mutation was noted on repeat liquid biopsy. Erlotinib, which had been discontinued while the patient was receiving chemotherapy was resumed after the completion of 6 cycles of carboplatin and etoposide, because of the possibility of heterogeneity of the tumor and documented persistence of a TKI-sensitive EGFR mutation.
NSCLC that harbors an EGFR-activating mutation usually carries a relatively good prognosis because it can be effectively treated with TKIs.4However, resistance to these therapies remains a persistent clinical problem. The most common resistance mechanism is the T790M mutation, reported to account for 50% to 60% of cases.5,6A less common form of resistance is transformation to SCLC, with various case reports observing this histologic transformation in 5% to 14% of patient biopsies at the time of TKI resistance.6,7Various genomic and molecular events have been described in the literature, which potentially could explain the transformation of NSCLC to SCLC, but the clonal evolution that underlies this transformation remains unclear.6,8Several studies have shown that the transformation is more common in EGFR-mutant adenocarcinoma than in EGFR wild-type adenocarcinoma.6,8
A major cause for transformation is thought to be genomic alteration. LADC with EGFR mutations that transform to SCLC after an initial response to EGFR TKI have been shown to retain EGFR mutation but develop additional loss or mutation in TP53 and RB1 genes, which are frequently seen in SCLC.6,8,9Additionally, the RNA profile of transformed LADC has been shown to mimic classical primary SCLC, while retaining a subset of mRNA more typically expressed in primary adenocarcinoma.10These findings support the hypothesis that these patients undergo a transformation of their primary adenocarcinoma, as opposed to having 2 primary lung cancers.6,8-10
The transformed SCLC is thought to have the same cells of origin as the adenocarcinoma. Historically, the cells of origin of SCLC and NSCLC are different. SCLC is thought to arise from neuroendocrine cells in the distal part of the airways, while adenocarcinoma is believed to arise from the alveolar type II cells located in the periphery.6One study in a mouse model showed that targeted deletion of TP53 and RB1 in alveolar type II cells led to the development of SCLC. The researchers concluded that alveolar type II cells might have the potential to give rise to SCLC, although at a lower frequency and after the deletion of TP53 and RB1 genes.8,9Thus, adenocarcinomas that develop from alveolar type II cells may have the potential to transform to SCLC if there is a deletion of TP53 and RB1 genes, suggesting that they may have developed from the same original clone.
Another hypothesis to explain this observation is that the patients may have had tumors with combined histology, which was initially not apparent. It should be noted that in reports that describe this observation, SCLC was dominant when the adenocarcinoma component was successfully treated with a TKI. However, rapid progression occurred in these cases and repeat biopsies showed classical SCLC with no adenocarcinoma component.6The results of several studies have shown that this observation is more common in EGFR-mutant adenocarcinoma than in EGFR wild-type adenocarcinoma.6,8
Once the tumor has transformed to SCLC, treatment with platinum and etoposide is commonly employed. This regimen is considered as the standard induction regimen for SCLC with reported response rates of 70% to 90% for limited-stage disease and 60% to 70% for extensive-stage disease.11Despite excellent response rates, most patients eventually progress, with dismal survival rates of 14 to 20 months for limited-stage disease and 9 to 11 months for extensive-stage disease.11No maintenance regimens have been shown to benefit patients with SCLC. For patients with transformed SCLC, there are no guidelines about whether to restart TKIs or not after induction chemotherapy. Tumor heterogeneity has been well documented, with small cell transformation occurring locally. Case reports of concurrent patchy small cell transformation and T790M resistance mutation in the same patients have been described.12
TKIs remain the standard of treatment for patients with EGFR-mutant LADC; however, resistance to TKIs is inevitable. Various forms of resistance have been described, but transformation to SCLC is less commonly encountered. Because of the increasing longevity of EGFR-mutated lung cancer patients, clinicians may see this transformation more often and its possibility should be borne in mind when patients exhibit progression on TKI treatment. The exact mechanism remains unclear. Changes in genomic/molecular characteristics, such as loss of or mutations in TP53 and RB1 genes, have been described in the literature.8,9Theories about common cellular origin and presence of combined histology at the time of diagnosis have also been described.6Immunohistochemistry assays for TP53 and RB1 genes may be utilized in the future to predict the chances of transformation to SCLC. However, further studies and investigations are necessary to identify the cause of this transformation and more accurately predict its risk.