Managing Patients With Relapsed Small-Cell Lung Cancer

Jun Gong; Ravi Salgia

doi:10.1200/JOP.18.00204

. 2018 Jun;14(6):359–366. doi: 10.1200/JOP.18.00204

Managing Patients With Relapsed Small-Cell Lung Cancer

Jun Gong ¹, Ravi Salgia ^1,^✉

PMCID: PMC6002253 PMID: 29894664

Abstract

Despite high response rates to initial therapy, relapses are common in patients with small-cell lung cancer (SCLC). Systemic therapy after first-line failure remains important in the treatment paradigm of SCLC. Reinitiation of a previously administered first-line chemotherapy regimen is recommended for relapse > 6 months from completion of initial therapy. For relapse ≤ 6 months of initial therapy, sequential therapy with single agents is recommended. Clinical trial enrollment should be considered at all stages of treatment of SCLC. This review highlights the available treatment options in relapsed SCLC. In particular, we focus on prospective clinical trials demonstrating activity for the most commonly used agents in this setting. We end with a discussion on future directions and emerging targets with potential to improve outcomes in relapsed SCLC.

INTRODUCTION

The incidence of small-cell lung cancer (SCLC) has steadily decreased in the United States, where it constitutes roughly 13% of all new lung cancer diagnoses.¹ In contrast to non–small-cell lung cancer, SCLC represents the most aggressive form of lung cancer, with a rapid doubling time, tendency for early and widespread metastases, and high sensitivity to initial therapy followed by eventual resistance on disease recurrence.² Response rates approximate 60% to 70% with initial therapy in limited-stage disease (disease for which the treatment standard remains chemotherapy with four to six cycles of cisplatin and etoposide and concurrent radiation therapy, followed by prophylactic cranial irradiation in responders).^2,3 For the 60% to 70% of patients presenting with extensive-stage disease at diagnosis, standard treatment includes four to six cycles of cisplatin or carboplatin and etoposide or irinotecan (clinical trials have not supported the superiority of irinotecan over etoposide in the United States and Europe), followed by consideration for prophylactic cranial irradiation and thoracic radiation therapy in responders.^2,4-6

The median overall survival (OS) approximates 20 months in limited-stage SCLC, with estimated 5-year survival rates of < 15%, whereas the median OS is approximately 8 to 12 months for extensive-stage SCLC, with 5-year survival rates of < 2%.⁷ Unfortunately, despite the rapid and impressive responses to initial therapy, relapse is often the rule rather than the exception; this places importance on salvage therapies for which many patients with SCLC are candidates for second-line therapy. In this review, we highlight available treatment options for relapsed or refractory SCLC, with particular emphasis on prospective clinical trials that have demonstrated activity for the most commonly used agents in this setting. We also provide recommendations on treatment approaches to SCLC after first-line treatment failure and discuss future considerations in optimizing therapeutic efficacy in relapsed SCLC.

RELAPSED OR REFRACTORY SCLC

Given the tendency to develop chemoresistance on disease recurrence, the hallmark of systemic therapy in SCLC after the patient experiences treatment failure with initial chemotherapy involves the introduction of non–cross-resistant chemotherapeutic agents. An important distinction recognized widely is the time from completion of initial chemotherapy to relapse: relapse ≤ 6 months versus relapse > 6 months, where reinitiation of the previously administered first-line chemotherapy regimen is recommended for SCLC relapse > 6 months.^4-6 Another popular distinction categorizes patients with SCLC who relapse > 3 months after first-line chemotherapy as chemosensitive, whereas those who relapse within 3 months of initial chemotherapy are considered chemoresistant; these latter patients with chemotherapy-refractory disease have < 10% response to subsequent chemotherapy.⁷ The overall response rate (ORR) and median progression-free survival (PFS) for patients with SCLC receiving second-line therapy is roughly 51% and 4.6 months, respectively.⁸ For patients with relapsed SCLC receiving third-line chemotherapy, the ORR is approximately 18%, and the median PFS is approximately 2.0 months.

Topotecan

The only US Food and Drug Administration (FDA)-approved agent in relapsed SCLC is topotecan (topoisomerase I inhibitor) on the basis of the phase III trial comparing intravenous topotecan with cyclophosphamide, doxorubicin, and vincristine (CAV) in relapsed sensitive disease, defined as progression ≥ 60 days after completion of first-line chemotherapy.⁹ Topotecan was shown to be as effective as CAV (Table 1); however, topotecan showed significant improvements across several disease-related symptoms over CAV. The FDA approval was subsequently expanded to include oral topotecan when proof of benefit over best supportive care and equivalent activity to intravenous topotecan in the second-line treatment of chemosensitive SCLC, defined as a treatment-free interval after initial therapy (TFI) ≥ 90 days, was demonstrated.^10,11 Notably, the superiority of topotecan over best supportive care was maintained, even in patients with a short TFI (≤ 60 days).¹⁰

Table 1.

Prospective Clinical Trials of Second-Line and Beyond Chemotherapy in Relapsed SCLC

Open in a new tab

Amrubicin

On the basis of promising activity in relapsed SCLC shown in phase II trials, second-line amrubicin (DNA topoisomerase II inhibitor) was investigated in the phase III ACT-1 trial against topotecan in chemorefractory (TFI < 90 days) or chemosensitive (TFI ≥ 90 days) SCLC.¹² Although amrubicin did not significantly improve OS over topotecan in the overall cohort, it did show a significant OS improvement over topotecan in the refractory SCLC subgroup (Table 1).

Irinotecan

A phase II trial was among the first to show the efficacy of single-agent irinotecan (weekly 100 mg/m²) as second-line therapy in relapsed sensitive (TFI > 90 days) SCLC (Table 1).¹³ The most common grade ≥ 3 adverse event was neutropenia (33%). The incidence of grade ≥ 3 diarrhea was 7%, and there were no patients with febrile neutropenia. A separate study showed the activity of second-line low-dose irinotecan (60 mg/m² weekly for 3 out of 4 weeks) in both chemorefractory (TFI ≤ 90 days) and chemosensitive (TFI > 90 days) SCLC.²¹ More recently, another phase II study reported an ORR of 41.3%, median PFS of 4.1 months, and median OS of 10.4 months with irinotecan (100 mg/m²) given every other week in 30 patients with relapsed SCLC.²² Notably, there are no phase III trials to date on irinotecan in relapsed SCLC.

Temozolomide

The oral alkylating agent temozolomide has demonstrated activity in the second- and third-line treatment of chemosensitive (TFI > 60 days) and chemoresistant (TFI ≤ 60 days) SCLC, with a noted 38% ORR in those with brain metastases and enhanced response in tumors that were O-6-methylguanine-DNA methyltransferase promoter methylated (38% v 7%; P = .08; Table 1).¹⁴ The same group later reported that temozolomide 200 mg/m²/d for 5 days every 28 days was the preferred schedule, given its tolerability and activity in a similar population of patients with relapsed SCLC.²³

Etoposide

An early phase II trial investigated single-agent oral etoposide 50 mg/m²/d every 21 days as second- and third-line treatment in chemosensitive (TFI > 90 days) or chemorefractory (TFI ≤ 90 days) relapsed SCLC (Table 1).¹⁵ Eighteen patients had received prior etoposide-based therapy, and 15 patients had prior cisplatin exposure, establishing this agent as an option in relapsed SCLC treated with conventional induction regimens.

Taxanes

In a phase II cohort of patients with SCLC relapsing within 3 months of the last cytotoxic therapy, paclitaxel showed an ORR of 29% in a heavily pretreated (up to three lines of prior therapy) and chemoresistant population (Table 1).¹⁶ Evidence exists to support a weekly paclitaxel regimen in relapsed SCLC as well.²⁴ A separate phase II trial reported an ORR of 25% for docetaxel in patients with relapsed SCLC who had previously received no more than one prior chemotherapy regimen.¹⁷

Gemcitabine

Single-agent gemcitabine produced a modest ORR of 13% in a phase II trial involving patients with SCLC who were treated with one or more lines of therapy and relapsed within 3 months after the last chemotherapeutic regimen or had primary progressive disease during the last regimen (Table 1).¹⁸ In a separate phase II trial, second-line gemcitabine showed an ORR of 11.9% and median OS of 7.1 months in a cohort including patients with chemosensitive SCLC (TFI ≥ 90 days).²⁵

Vinca Alkaloids

A phase II trial showed efficacy for second-line vinflunine in relapsed chemosensitive (TFI > 3 months) and chemorefractory (progressive disease during or relapse within 3 months of initial therapy) SCLC (Table 1).¹⁹ Before this study, weekly vinorelbine had produced a modest response rate (12.5%) in a heavily pretreated relapsed or refractory SCLC cohort in the phase II setting.²⁶

Bendamustine

Bendamustine is another alkylating agent that has demonstrated activity in platinum-sensitive (TFI ≥ 90 days) or platinum-resistant (TFI < 90 days) SCLC in patients who had previously received up to two lines of prior therapy (Table 1).²⁰ Notably, this study failed to meet its prespecified primary end point of a median time to progression of 4.5 months.

Checkpoint Inhibitors

The response of the cohort with relapsed SCLC from the phase I/II CheckMate 032 trial enrolling patients with platinum-sensitive (relapse ≥ 90 days) and platinum-resistant (relapse < 90 days or progressive disease during chemotherapy) disease after one or more previous platinum-containing regimens to the programmed cell death protein-1 inhibitor nivolumab with or without the cytotoxic T-lymphocyte–associated protein 4 inhibitor ipilimumab was recently reported (Table 2).²⁷ Nivolumab monotherapy and nivolumab plus ipilimumab showed activity and durable responses regardless of programmed cell death ligand-1 expression. Grade 3 to 4 treatment-related adverse events were lowest (19%) in the nivolumab 3 mg/kg and ipilimumab 1 mg/kg arm and highest (30%) in the nivolumab 1 mg/kg and ipilimumab 3 mg/kg arm.

Table 2.

Recent Clinical Trials of Novel Combinations and Agents in Relapsed SCLC

Open in a new tab

Preliminary results from the phase Ib KEYNOTE-028 reported durable responses ≥ 16 weeks to single-agent pembrolizumab in 25% of patients with SCLC previously treated with platinum and etoposide (Table 2).²⁸ Tumors in treated patients must have had programmed cell death ligand-1 expression ≥ 1%, and thus far, only one patient has had a grade ≥ 3 treatment-related adverse event.

Combination Chemotherapy

Multiagent chemotherapy has historically demonstrated higher response rates compared with single-agent chemotherapy in relapsed SCLC, although often at the expense of increased toxicities.^32,33 A recent Japanese phase III trial demonstrated the superiority of the combination of cisplatin, irinotecan, and etoposide over topotecan alone in the second-line treatment of sensitive relapsed SCLC (TFI ≥ 90 days; Table 2).²⁹ Notably, grade 3 to 4 febrile neutropenia was higher in the combination arm (31%) than the topotecan arm (7%), with serious adverse events seen in 10% of the combination arm compared with 4% in the topotecan arm.

Targeted Therapy

Initial investigations of targeted agents have largely failed to demonstrate proven efficacy in relapsed SCLC and have been extensively reviewed elsewhere.⁷ Recently, a phase II trial demonstrated the efficacy of the tyrosine kinase inhibitor pazopanib in the second-line treatment of sensitive (TFI > 90 days) and resistant (TFI ≤ 90 days) SCLC (Table 2).³⁰ In a multicenter phase I/II trial, the aurora kinase A inhibitor alisertib produced an ORR of 21% in a refractory SCLC cohort including 48 patients with resistant (TFI ≤ 90 days) or sensitive (TFI > 90 days) disease (Table 2).³¹ The most common grade 3 to 4 treatment-related adverse event was neutropenia (37%). Rovalpituzumab is an antibody-drug conjugate directed against the Notch ligand delta-like protein 3 and produced an ORR of 18% in 60 patients with relapsed SCLC or large-cell neuroendocrine carcinoma.³⁴ After these findings, however, rovalpituzumab was not moved forward to later-phase clinical trials. Despite the results of these more recent trials, there has yet to be an established molecularly targeted therapy resulting in FDA approval for the treatment of SCLC.⁴

DISCUSSION

Given that relapses are common for the majority of patients with SCLC, systemic therapy after first-line treatment failure remains an important component of the treatment paradigm for SCLC. In an initial single-institution experience, 35 of 191 patients with SCLC (18%) received third-line chemotherapy with a median time from diagnosis to start of third-line treatment of 15 months (range, 5 to 34).³⁵ A separate multicenter retrospective analysis included 120 patients who received third-line chemotherapy for relapsed SCLC from 2000 to 2010; rechallenge with platinum-based chemotherapy (57%) was the most common second-line regimen, followed by CAV (26%), enrollment in a clinical trial (17%), and topotecan (11%).⁸ The most common third-line regimens were CAV (43%), platinum-based chemotherapy (24%), and topotecan (17%). Patients who received third-line chemotherapy tended to be younger, have a lower performance status, and have better response to treatment in earlier-line settings.³⁵

The decision for second-line and beyond treatment in relapsed SCLC should take into consideration the performance status and comorbidities of the patient, organ reserve, toxicity experienced from previous chemotherapy, response and disease-free interval from prior therapy, and treatment goals. All patients with relapsed SCLC should be considered for clinical trials (Fig 1). In patients with relapse > 6 months from completion of initial chemotherapy, retreatment with initial chemotherapy remains an important consideration, and these patients should be rechallenged with first-line chemotherapy.^4-6 For patients with relapse ≤ 6 months from first-line therapy, we recommend sequential, single-agent therapy, in keeping with several major guidelines.^4-6 With the exception of CAV, which has been shown to be as effective as topotecan, and the recently investigated CIE regimen, most regimens of combination chemotherapy in relapsed SCLC have been investigated in phase II settings.³³ Multiagent chemotherapy has been associated with higher response rates than single-agent therapy in relapsed SCLC but at the expense of heightened toxicities.^32,33 In a meta-analysis of 66 phase II/III trials involving 8,471 patients with SCLC, there was no significant difference in outcomes (ORR, PFS, or OS) between multiple-arm and single-arm trials.²⁹ Furthermore, multiagent chemotherapy did not reverse the likelihood of response in the second-line treatment of resistant/refractory SCLC.³³

Fig 1. — Treatment algorithm for relapsed small-cell lung cancer (SCLC). CAV, cyclophosphamide, doxorubicin, and vincristine; CIE, cisplatin, irinotecan, and etoposide; TFI, treatment-free interval after initial therapy. (*)Enrollment into clinical trials should be considered at all stages of treatment of SCLC. (†)For programmed cell death ligand-1 expression ≥ 1%. (‡)For second-line treatment of sensitive relapsed SCLC (TFI ≥ 90 days).

It should be noted that the recent phase III trial demonstrating the superiority of second-line CIE over topotecan in sensitive relapsed SCLC enrolled a Japanese population where efficacy results have historically fared better than patients in US and European trials; it used a lower topotecan dose (1.0 mg/m²) than the FDA-approved dose (1.5 mg/m²), did not include quality-of-life end points, and subsequent therapy with three or more drugs was prohibited in the topotecan arm, which could have contributed to the difference in outcomes seen between the combination chemotherapy and topotecan arms.²⁹ We have included this as a treatment option in relapsed SCLC, although with the understanding that incorporation into routine clinical practice has not been as widespread as other second-line regimens. At our institution, we have prioritized clinical trials, and from a standard-of-care perspective, we have favored the use of programmed cell death protein-1 inhibitors, in particular, pembrolizumab or nivolumab, as the preferred second-line regimen on the basis of their efficacy and tolerability in relapsed SCLC (Table 2). Admittedly, confirmation in later-phase trials with immunotherapy in relapsed SCLC is eagerly awaited to establish this class of agents in the second-line and beyond armamentarium.

Large analyses of phase II/III SCLC trials have unfortunately demonstrated that over a 27- to 28-year period, no significant improvement in OS over time has been achieved in SCLC.^32,36 Few advancements have been made in the systemic treatment of SCLC since the last drug approved by the FDA, topotecan, nearly two decades ago. However, recent developments in the preclinical and clinical settings offer enthusiasm for the potential to advance the field in the management of SCLC. The first comprehensive genomic profiling of 110 SCLC tumors from patients with predominantly early-stage and treatment-naïve disease identified several recurrent somatic alterations involving potentially targetable signaling pathways, such as cell cycle regulation (CDKN2A, TP53, CCND1, RBL1, RBL2, RB1, TP73), receptor kinase/PI3K signaling (KIT, FGFR1, IRS2, PIK3CA, PTEN), chromatin remodeling and cellular proliferation (CREBBP, EP300, MYC), and Notch signaling/neuroendocrine differentiation (NOTCH).³⁷

Targeting of several of these pathways has already undergone preclinical and early-phase clinical development. Targeting of MYC and MYC-related aurora kinase signaling in SCLC has already demonstrated feasibility in preclinical and early-phase settings.^31,38 Combined inhibition of PARP and PI3K/mammalian target of rapamycin showed enhanced antitumor activity in animal models of SCLC.³⁹ SCLC cell lines have shown sensitivity to FGFR1 inhibition that may correlate with increased FGFR1 mRNA expression.⁴⁰ Novel alterations in c-MET, including missense mutations leading to exon 14 skipping identified in SCLC cell lines, contribute to tumorigenicity and can serve as potential therapeutic targets.⁴¹ In addition, targeting of pathways involved in DNA methylation (EZH2), cytokine and tyrosine kinase signaling (JAK, RET, c-Kit, RON), apoptosis (BCL2), differentiation and cellular proliferation (Hedgehog, PAX5), chromatin modification (bromodomain, LSD1), and DNA repair (WEE1, CHK1) has begun investigation and is at various stages of development in SCLC.^42-48 The search for predictive biomarkers in SCLC is also picking up momentum. Response to treatment with a panel of 103 FDA-approved and 423 investigational agents in SCLC cell lines has been associated with global gene and microRNA expression profiles that are publicly available.⁴⁹ Examination of copy-number alterations in circulating tumor cells from pretreatment SCLC blood samples has identified copy-number alteration profiles associated with chemoresistance and chemosensitivity that may have relevance in the relapsed setting.⁵⁰

In conclusion, systemic therapy after first-line treatment failure remains an important component of the treatment paradigm for SCLC, given that most patients will experience relapse during the course of the disease. For relapse > 6 months from completion of initial therapy, reinitiation of the previously administered first-line chemotherapy regimen is recommended. For relapse ≤ 6 months from initial therapy, sequential therapy with single agents is recommended. Enrollment in clinical trials is an important consideration at all stages of treatment of SCLC. Results from larger-scale, confirmatory clinical trials involving checkpoint inhibitors and other novel agents are eagerly anticipated in relapsed SCLC. Concurrent development of predictive biomarkers will hopefully lead to improvement in outcomes in relapsed SCLC as well.

ACKNOWLEDGMENT

Research reported in this article was supported by the National Cancer Institute of the National Institutes of Health under Grants No. P30CA033572 and 1U54CA209978-01A1. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

AUTHOR CONTRIBUTIONS

Conception and design: Ravi Salgia

Collection and assembly of data: All authors

Data analysis and interpretation: All authors

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST