etoposide in extensive‐disease small cell lung cancer

Lung Cancer 105 (2017) 7–13

Contents lists available at ScienceDirect

Lung Cancer journal homepage: www.elsevier.com/locate/lungcan

A randomized phase II study of LY2510924 and carboplatin/etoposide versus carboplatin/etoposide in extensive-disease small cell lung cancer夽 Ravi Salgia a,∗ , John R. Stille b , R. Waide Weaver c , Michael McCleod d , Oday Hamid b , John Polzer b , Stephanie Roberson b , Amy Flynt e , David R. Spigel f a

Department of Medicine, The University of Chicago, Chicago, IL, United States Eli Lilly and Company, Indianapolis, IN, United States c Florida Cancer Specialists, St. Petersburg, FL, United States d Florida Cancer Specialists, Fort Myers, FL, United States e PharPoint Research Inc., Durham, NC, United States f Lung Cancer Research Program, Sarah Cannon Research Institute, LLC, Nashville, TN, United States b

a r t i c l e

i n f o

Article history: Received 10 June 2016 Received in revised form 15 December 2016 Accepted 26 December 2016 Keywords: CXCR4 LY2510924 Extensive-disease small cell lung cancer Phase II trial

a b s t r a c t Objectives: This multicenter, open-label, randomized phase II study evaluated the efficacy and safety of LY2510924 (LY) added to first-line standard of care (SOC) chemotherapy for extensive-disease small cell lung cancer (ED-SCLC) and explored the predictive value of C-X-C motif receptor 4 (CXCR4) tumor response. Materials and methods: Patients with treatment-naïve ED-SCLC were randomized (1:1) to receive up to six 21-day cycles of carboplatin/etoposide alone (SOC) or in combination with 20 mg LY2510924 administered subcutaneously on days 1–7 of each cycle (LY + SOC). The primary efficacy endpoint was progression-free survival (PFS). Secondary endpoints were overall survival (OS), overall response rate (ORR), and safety. Response relative to CXCR4 expression on baseline tumor was an exploratory endpoint. Results: Of 94 patients randomized, 90 received treatment (LY + SOC, n = 47; SOC, n = 43). Median PFS (95% confidence interval [CI]) was 5.88 (4.83, 6.24) months for LY + SOC versus 5.85 (4.63, 5.51) months for SOC (hazard ratio [95% CI], 1.01 [0.62, 1.63]; p = 0.9806). Median OS (95% CI) was 9.72 (6.64, 11.70) months for LY + SOC versus 11.14 (8.25, 13.44) months for SOC. ORR was 74.5% for LY + SOC versus 81% for SOC. Safety results between arms were similar, although the following adverse events were more frequent on the LY + SOC arm: anemia (61.7% vs 46.5%), neutropenia (61.7% vs 53.5%), leukopenia (27.7% vs 9.3%), vomiting (27.7% vs 16.3%), and pneumonia (10.6% vs 2.3%). In patients whose baseline CXCR4 expression was above the optimal cutoff (H-score 210), the hazard ratio (95% CI) was 1.27 (0.51, 3.15). Conclusion: LY2510924 did not improve efficacy but had an acceptable toxicity profile when added to SOC for ED-SCLC. © 2017 Elsevier B.V. All rights reserved.

1. Introduction Approximately 13% of the estimated 224,000 new lung cancers diagnosed in the United States annually are expected to be small cell lung cancers (SCLCs) [1]. The majority (65%–70%) of these are expected to be extensive-disease SCLCs (ED-SCLCs).

夽 Trial registration ID: NCT01439568. ∗ Corresponding author (present address) at: Department of Medical Oncology and Therapeutics Research, City of Hope, 1500 E. Duarte Road, Duarte, CA 910103000, United States. E-mail address: [email protected] (R. Salgia). http://dx.doi.org/10.1016/j.lungcan.2016.12.020 0169-5002/© 2017 Elsevier B.V. All rights reserved.

The current standard of care (SOC) for ED-SCLC is combination of a platinum agent (cisplatin or carboplatin) and etoposide. Over the last 25 years, changes in regimen, dose intensity, schedule, and duration have only minimally improved survival [2,3]. A chemotherapeutic response is frequently achieved, but median overall survival (OS) remains only 8–10 months. The chemokine (C-X-C motif) receptor 4 (CXCR4) and its only known ligand, CXCL12 (also known as ␣-chemokine stromal cellderived factor 1 [SDF-1]), appear to play an important role in the regulation of organ-specific metastasis as well as tumor growth, invasion, survival, and angiogenesis. CXCR4 is functionally expressed or overexpressed in a variety of solid tumor cancers

8

R. Salgia et al. / Lung Cancer 105 (2017) 7–13

and in lymphoma and chronic lymphocytic leukemia [4,5]. Within hypoxic tumor tissue, observed increases in both SDF-1 secretion by fibroblasts and CXCR4 expression on tumor cells appear to stimulate tumor cell growth, migration, and invasion. SDF-1 promotes tumor growth in paracrine fashion by direct stimulation of tumor cell proliferation and survival via CXCR4. Stromal cells in tissues such as bone, liver, and lung secrete SDF-1, thereby creating a concentration gradient that induces directional migration of CXCR4-expressing cancer cells toward those tissues [6]. LY2510924 is a potent, selective peptide antagonist of CXCR4 in vitro and in vivo. LY2510924 inhibits SDF-1 binding to CXCR4 from different species, including that in human, monkey, mouse, and rat cells [7]. This molecule also inhibits SDF-1␣ binding to human CXCR4 in concentration-dependent fashion. In a phase 1 trial, LY2510924 was well tolerated at a maximum tolerated dose of 20 mg administered subcutaneously once daily [8]. CXCR4 is ubiquitously expressed in SCLC cells [9]. CXCR4 activation induces migration and invasion into the extracellular matrix (ECM) and adhesion to marrow stromal cells in CXCR4- and integrin-dependent fashion. Thus, CXCR4 may direct the distinct marrow-directed pattern of metastasis observed in patients with SCLC. Adhesion to marrow stromal cells or ECM proteins via integrins can protect SCLC cells from chemotherapy-induced apoptosis, which can lead to drug resistance, tumor cell growth, and residual disease and the relapse commonly seen in patients with SCLC. Consequently, CXCR4 overexpression in SCLC has been described as a prognostic biomarker [10,11]. However, stromal cell protection of SCLC cells has been inhibited by experimental CXCR4 antagonists such as T-140 [12]. Here, we report the results of a multicenter, open-label, randomized phase II study that compared the efficacy, safety, pharmacokinetics, and pharmacodynamics of the CXCR4 antagonist LY2510924 when added to first-line SOC chemotherapy versus SOC alone in patients with ED-SCLC. 2. Patients and methods 2.1. Patients Eligible patients were ≥18 years old and had histologically or cytologically confirmed ED-SCLC, no prior chemotherapy for SCLC, an Eastern Cooperative Oncology Group (ECOG) performance status of ≤2, and adequate organ function (see Supplementary Materials for main inclusion/exclusion criteria). All enrolled patients provided written informed consent approved by local independent ethics committees. The trial was registered at clinicaltrials.gov (NCT01439568). 2.2. Study design and treatment The primary objective was to compare the progression-free survival (PFS) between treatment arms. Secondary objectives were OS, overall response rate (ORR), duration of overall response (DOR), and safety. Patients were randomly assigned (1:1) to receive up to six 21day cycles of carboplatin/etoposide alone (SOC) or in combination with 20 mg LY2510924 administered subcutaneously on days 1-7 of each cycle (LY + SOC). In both arms, SOC included standard doses of carboplatin (5 areas under the curve [AUCs] intravenous [iv]) on day 1 and etoposide (100 mg/m2 iv) on days 1-3 of each cycle. In the LY + SOC arm, LY2510924 20 mg was injected subcutaneously once daily. All patients could receive full supportive care on study, including granulocyte and erythrocyte growth factors, but no other cancer treatment. Per standard practice, patients with CNS metastasis who showed response after 6 cycles of therapy were eligible to receive

prophylactic cranial irradiation administered according to local institutional guidelines. 2.3. Assessments 2.3.1. Baseline Pretreatment baseline procedures included complete medical history, physical examination, ECOG performance status evaluation, tumor staging, and laboratory tests (chemistry, hematology, and urinalysis). All staging procedures, including computed tomography (CT) or magnetic resonance imaging (MRI) of head, chest, abdomen, and pelvis and a bone and/or positron emission tomography scan, were done within 28 days of treatment initiation. 2.3.2. Efficacy PFS was defined as the time from randomization to disease progression or death from any cause. For patients not known to have died at the data cutoff date and who had no objective progressive disease, PFS was censored at the date of the last objective progression-free disease assessment. OS was defined as the time from study enrollment to death from any cause. Disease recurrence was evaluated per Response Evaluation Criteria In Solid Tumors (RECIST) v1.1. All patients underwent repeat CT/MRI after Cycles 2, 4, and 6 and then every 6-8 weeks until disease progression, death, or study closure. 2.3.3. Safety Adverse events (AEs), toxicities, ECGs, and clinical laboratory test results were assessed. All AEs occurring during study treatment or within 30 days of the last dose of study treatment, regardless of causality, were graded using the Common Terminology Criteria for Adverse Events, version 4.0. 2.3.4. Biomarkers 2.3.4.1. Peripheral biomarkers. Sampling for CD34+ stem cell analysis occurred predose and 4–6 h postdose on cycle 1 day 1 and cycle 2 day 1. Predose sampling for SDF-1 and interleukin 6 (IL6) occurred on cycle 1 days 1 and 7 and cycle 2 day 1. Sampling for IL-6 analysis occurred 4-6 h postdose on cycle 1 days 1 and 7. 2.3.4.2. Baseline CXCR4 expression in tumors by immunohistochemistry. CXCR4 expression analysis was performed on formalin-fixed paraffin-embedded (FFPE) tumor tissue biopsies taken from the diagnosis tissue samples, when available, before treatment initiation. Immunohistochemistry (IHC) staining for CXCR4 (rabbit anti-human CXCR4 monoclonal antibody, clone UMB2; Epitomics Inc., Burlingame, CA) was performed at Bostwick Laboratories (Uniondale, NY). IHC staining was performed on an intelliPATH FLX Autostainer (Biocare Medical, Concord, CA) using heat-induced antigen retrieval and standard techniques with 3,3 -diaminobenzidine as chromogen and counterstained with hematoxylin. Stained slides were reviewed by board-certified pathologists and assigned an H-score (0-300) for CXCR4 expression. Immunostaining intensity was visually assessed as 0 (no staining), 1+ (low intensity), 2+ (intermediate intensity), or 3+ (strong intensity). H-score was calculated from the percentage of cells (in 10% increments) at different staining intensities as follows: 0 (% 0 staining cells) + 1 × (% 1+ staining cells) + 2 × (% 2+ staining cells) + 3 × (% 3+ staining cells). 2.4. Statistical analysis methods Computer-generated randomization occurred before but as close as possible to treatment start, with stratification by ECOG per-

R. Salgia et al. / Lung Cancer 105 (2017) 7–13

9

Fig. 1. Patient disposition.

formance status (0/1 vs 2) and serum lactate dehydrogenase level (≤1× upper limit of normal [ULN] vs >1 ULN) at baseline. The sample size of 90 randomized patients was chosen to allow observance of 66 PFS events. This number of PFS events was calculated to provide 70% power to detect a 40% increase in PFS (from 5.2 months to 7.3 months) and at least 80% power to detect a 51% increase in PFS (from 5.2 months to 7.9 months), using a log-rank test with a 1-sided 0.2 type I error level. For PFS and OS analysis, Kaplan-Meier curves, hazard ratios (HRs), 95% confidence intervals (CIs), and stratified log-rank test p-values were calculated. For DOR analysis, HR, 95% CI, and stratified log-rank test p-value were calculated. For ORR analysis, ORRs and 95% CIs were estimated using unadjusted normal approximation for binomial proportions (z approximation); although planned, ORR was not compared statistically between treatment groups. All efficacy analyses used the efficacy population (i.e., all randomized patients who received at least 1 dose of study drug, with the exception of 1 patient randomized to the SOC arm but excluded because of a major protocol violation). All safety analyses used the safety population (i.e., all randomized patients who received at least 1 dose of study drug). Exploratory graphical analysis of peripheral biomarkers CD34+, SDF-1, and IL-6 was conducted to evaluate biomarker response to treatment. Receiver operating characteristic (ROC) curves, AUCs, and 95% CIs were generated to determine the optimal cutoff value for CXCR4 expression on tumor tissue at baseline as a predictor of 6-month PFS. The sensitivity, specificity, positive predictive value, and negative predictive value based on the optimum cutoff value were calculated.

2011, and the last patient completed the study on 14 August 2013. Ninety patients received treatment (see Fig. 1). Similar percentages of patients on the LY + SOC and SOC arms completed 6 treatment cycles (61.7% vs 60.5%). The median number of cycles completed was 6 on both arms. Treatment discontinuation due to progressive disease occurred more frequently on the SOC arm (18.6% vs 8.5%).

3. Results

3.4. Safety

3.1. Patients and treatment

The most frequent AEs by arm are shown in Supplementary Table 1. The following AEs had incidence rates more than 8% higher on the LY + SOC arm than the SOC arm: anemia (61.7% vs 46.5%), neutropenia (61.7% vs 53.5%), leukopenia (27.7% vs 9.3%), vomiting

Ninety-four patients were randomized to treatment at 16 US sites. The first patient signed informed consent on 22 September

3.2. Baseline demographic and clinical characteristics Both treatment arms were similar in terms of most baseline demographic and clinical characteristics (Table 1), though the SOC arm had proportionally more females. Tumor CXCR4+ expression assessed by IHC at baseline and mean and median H-scores were comparable between arms. 3.3. Efficacy Median PFS (95% CI) did not differ between treatment arms: 5.88 (4.83, 6.24) months for LY + SOC versus 5.85 (4.63, 6.51) months for SOC (HR [95% CI], 1.01 [0.62, 1.63]; p = 0.9806, 2-sided log-rank test) (Fig. 2A). For patients with an H-score ≥210 for CXCR4 expression on tumor tissue, the HR (95% CI) for PFS between the LY + SOC and SOC arms was 1.27 (0.51, 3.15) (Fig. 3). Median OS (95% CI) was 9.72 months (6.64, 11.70) for LY + SOC and 11.14 months (8.25, 13.44) for SOC (HR [95% CI], 1.51 [0.90, 2.53]; p = 0.1120, 2-sided log-rank test) (Fig. 2B). ORR was 74.5% [35/47] for LY + SOC and 81% [34/42] for SOC (Table 2). Median DOR was also similar between arms (4.83 months [LY + SOC] vs 4.67 months [SOC]) (HR [95% CI], 1.00 [0.55, 1.80]; p = 0.9985, 2-sided log-rank test).

10

R. Salgia et al. / Lung Cancer 105 (2017) 7–13

Fig. 2. Progression-free survival (A) and overall survival (B) in efficacy population.

(27.7% vs 16.3%), and pneumonia (10.6% vs 2.3%). Most AEs were grade 1 or 2 (see Supplementary Table 2). The most frequent grade 3/4 AEs by arm are shown in Table 3. The following were more frequent on the LY + SOC arm: platelet count decreased (23.4% vs 16.3%), lung infection (10.6% vs 0%), and febrile neutropenia (8.5% vs 2.3%). During the treatment period, there was 1 death not due to progression of disease (ruptured intracerebral aneurysm on the SOC arm). During the 30-day follow-up period after completing the last treatment cycle, there were 3 deaths not due to disease progression (myocardial infarction and cause unknown on the LY + SOC arm and cardiac arrest on the SOC arm). Serious adverse events (SAEs) occurred more frequently on the LY + SOC arm (51% [24/47] vs 30.2% [13/43]). Two patients on the LY + SOC arm experienced SAEs considered possibly related to LY2510924 treatment (all grade 3), including febrile neutropenia

and lower gastrointestinal hemorrhage in 1 patient and anemia in another. Treatment discontinuation due to progressive disease occurred more frequently on the SOC arm (18.6% vs 8.5%). Conversely, treatment discontinuation due to AEs occurred more frequently on the LY + SOC arm (17.0% vs 4.7%) (Supplementary Table 3). On the LY + SOC arm, this included 8 patients who experienced neutropenia (2 patients), exacerbation of chronic obstructive pulmonary disease (COPD), sepsis, myocardial infarction, fracture, hyponatremia, and pneumonia. On the SOC arm, this included 2 patients who experienced neutropenia and intracerebral aneurysm. Of these AEs leading to treatment discontinuation, only the COPD (LY + SOC) and the neutropenia (SOC) were considered possibly related to study treatment; the neutropenia eventually resolved, but the COPD did not.

R. Salgia et al. / Lung Cancer 105 (2017) 7–13

11

Fig. 3. Progression–free survival for patients with CXCR4+ H-score ≥210 on tumor tissue in efficacy population. Also shown are associated hazard ratio (HR) and 95% confidence interval (CI) through 4 months.

Table 2 Best overall response (efficacy population).

Table 1 Patient demographics and disease characteristics at baseline. LY + SOC (N = 47)

SOC (N = 43)

Total (N = 90)

Sex, n (%) Male Female

22 (46.8) 25 (53.2)

17 (39.5) 26 (60.5)

39 (43.3) 51 (56.7)

Age (years) Mean (SD) Minimum, maximum Median

63.69 (9.033) 67.07 (8.328) 65.31 (8.820) 46.5, 84.1 47.8, 85.4 46.5, 85.4 65.30 66.10 65.95

Race, n (%) American Indian/Alaskan Native Asian Black/African American White Unknown Other

1 (2.1) 0 3 (6.4) 43 (91.5) 0 0

0 0 2 (4.7) 40 (93.0) 0 1 (2.3)

1 (1.1) 0 5 (5.6) 83 (92.2) 0 1 (1.1)

Characteristic

Stage IV at initial diagnosis, n (%)

47 (100.0)

43 (100.0)

90 (100.0)

ECOG PS, n (%) 0 1 2 3 4 5

47 (100.0) 16 (34.0) 28 (59.6) 3 (6.4) 0 0 0

43 (100.0) 15 (34.9) 25 (58.1) 3 (7.0) 0 0 0

90 (100.0) 31 (34.4) 53 (58.9) 6 (6.7) 0 0 0

CXCR4+ on tumor tissue by IHCa , b Mean H-score (SD) Median H-score (range)

206 (80) 200 (85) 200 (20–200) 200 (5–300)

– –

ECOG PS: Eastern Cooperative Oncology Group performance status; IHC: immunohistochemistry; LY + SOC: LY2510924 (20 mg) + carboplatin/etoposide; SD: standard deviation; SOC: carboplatin/etoposide. a H-scores were calculated as described in Results section. b No. patients with evaluable results: LY + SOC (n = 36) and SOC (n = 33).

3.5. Biomarkers 3.5.1. Peripheral biomarkers LY + SOC treatment led to a posttreatment increase in absolute CD34+ stem cell count, which confirmed pharmacodynamic response to inhibition of the CXCR4-SDF-1 axis (Supplementary Fig. 1). In contrast, SOC treatment led to a slight decrease in absolute CD34+ stem cell count. During treatment holiday (days 8-21),

ORR (CR + PR) Complete response Partial response Stable disease Progressive disease Not evaluable Missing

LY + SOC (N = 47)

SOC (N = 42)

35 (74.5) 1 (2.1) 34 (72.3) 6 (12.8) 1 (2.1) 1 (2.1) 4 (8.5)

34 (81.0) 2 (4.8) 32 (76.2) 3 (7.1) 1 (2.4) 0 (0.0) 4 (9.5)

CR: complete response; PR: partial response; LY + SOC: LY2510924 (20 mg) + carboplatin/etoposide; SOC: carboplatin/etoposide. Note: Data are presented as n (%) of patients.

Table 3 Grade 3/4 treatment-emergent adverse events reported by >5% of patients on LY + SOC arm (safety population).

Laboratory Neutrophil count decreased Anemia Platelet count decreased White blood cell count decreased Non-laboratory Lung infection Febrile neutropenia

LY + SOC (N = 47)

SOC (N = 43)

19 (40.4) 14 (29.8) 11 (23.4) 4 (8.5)

24 (55.8) 14 (32.6) 7 (16.3) 4 (9.3)

5 (10.6) 4 (8.5)

0 1 (2.3)

LY + SOC: LY2510924 (20 mg) + carboplatin/etoposide; SOC: carboplatin/etoposide. Note: Data are presented as n (%) of patients.

absolute CD34+ stem cell count increased on both treatment arms. Analyses of SDF-1 and IL-6 levels did not reveal any differences between arms (data not shown).

3.5.2. Optimal cutoff value for CXCR4 expression on tumor tissue The optimal ROC cutoff value for CXCR4 expression on tumor tissue as a predictor of 6-month PFS was an H-score of 210. Overall, H-score was low (< 210) in 38 patients and high (≥210) in 31 patients, resulting in a sensitivity of 54.8% and specificity of 70.4%.

12

R. Salgia et al. / Lung Cancer 105 (2017) 7–13

4. Discussion ED-SCLC is difficult to treat. Platinum therapies plus etoposide have been used for decades and remain the standard treatment with median survival of 8–10 months [13]. Strategies for improving survival have included adding a third agent to the current platinum-based regimen [14], intensifying dose [15], alternating non-cross-resistant regimens [16], using other regimens as alternatives to standard platinum-based therapy [17], and using maintenance chemotherapy [3]. None of these strategies has significantly improved survival. Most patients initially present with metastatic disease. There are various mechanisms of metastasis in SCLC, including activation of receptor tyrosine kinases such as MET/RON [18], modulation of downstream signal transduction pathways, and activation/inactivation of genetic pathways. CXCR4 plays an important role in the signal transduction of various pathways and also in the migration/mobility of SCLC cells [19]. Previously, we had shown that CXCR4 can be activated by its ligand, CXCL12, in a number of SCLC cell lines and activate downstream pathways of AKT, thus improving survival [20]. Here, we examined whether CXCR4 inhibition (via LY2510924) could be combined with standard cytotoxic chemotherapy. In this randomized phase II study, addition of CXCR4 inhibitor LY2510924 to first-line standard platinum-based chemotherapy did not improve efficacy (as measured by median PFS, OS, ORR, or DOR) in treatment-naive ED-SCLC patients. Patients whose tumors overexpressed CXCR4 (H-score ≥ 210) did not appear to benefit more from LY2510924 plus chemotherapy than from chemotherapy alone. However, in an exploratory analysis of PFS in such patients, the hazard ratio was numerically better for the combination than for chemotherapy alone through 4 months (end of study treatment) with an HR (95% CI) of 0.79 (0.21, 2.93) (Fig. 3). Several trials against the CXCR4/CXCL12 axis in hematologic malignancies are ongoing, including trials of BL-8040 and BMS 936564 (see clinicaltrials.gov). Plerixafor appears to have potential efficacy in Waldenström’s macroglobulinemia with WHIM mutations in CXCR4 [21]. However, in solid tumors, this axis inhibition has been minimally explored. Two studies, a phase 1 study of plerixafor and bevacizumab in patients with recurrent high-grade gliomas and a phase II study of sunitinib and LY2510924 in renal cell carcinoma, have shown disappointing efficacy results, although the combination used in each study was well tolerated [22,23]. Several factors may explain the inability to see a difference between the 2 treatment arms. First, ED-SCLC may be too far advanced for CXCR4 inhibitors to work. Limited-disease SCLC may be a more appropriate disease in which to investigate this agent. Preventing tumor cell mobilization may be a way to prevent limited disease from advancing to extensive disease. Inhibition of CXCR4 deters migratory and invasive cells from colonizing stromal structures and therefore theoretically limits their circulation. Second, proportionally more patients discontinued due to an AE on the LY + SOC arm than on the SOC arm (17.0% vs 4.7%), while more patients discontinued due to PD on the SOC arm than on the LY + SOC arm (18.6% vs 8.5%), theoretically suggesting that improving tolerability may lead to improvement in PFS. Third, the therapy against CXCR4 may not work in SCLC, thus making SCLC a poor target. Perhaps only a subset of patients would benefit from this combination therapy, which makes identifying a potential predictive biomarker that would lead to a positive outcome for such a trial important. Even though the CXCR4/CXCL12 pathway has been shown to be functional in SCLC cell lines, this pathway may need to synergize with other molecules such as KIT, as we had shown initially [9], or integrins. As shown in certain head and neck cancers and Waldenström’s macroglobulinemia, there are WHIM mutations in CXCR4 [21]; however, there are no mutations

of CXCR4 in SCLC (unpublished observations). Since there are no activating mutations of CXCR4, this may not lead then to activation in SCLC. Predictive biomarkers are integral to non-SCLC therapeutics. For example, patients with EML4-ALK or ROS1 translocations can respond to targeted therapeutics such as crizotinib [24]. In SCLC, however, there are a number of potentially prognostic biomarkers (e.g., CXCR4) being explored, but predictive biomarkers are desperately needed. There is evidence that phosphorylated topoisomerase I may be a predictive biomarker for response to anti-topoisomerase I strategies [25]. We sought to determine if there could be predictive value to CXCR4+ , but could demonstrate none. Meanwhile, other potential “liquid” biomarkers need to be explored, such as circulating tumor cells, cell-free DNA, and exosomes. Although the present study did not meet its primary efficacy endpoint, we did observe that LY2510924 is safe when given in combination with carboplatin and etoposide. The majority of side effects were as expected for the combination cytotoxic chemotherapy. In summary, LY2510924 in combination with carboplatin/etoposide chemotherapy was relatively safe and tolerable in patients with ED-SCLC. There was no observed difference in PFS and OS between treatment arms, and without an established biomarker, this therapeutic approach should not be pursued further. However, as the genetics of SCLC become clearer [26], we should determine further the combinations of CXCR4 inhibition and other novel pathways that might be useful in the treatment of SCLC. Author contributions The sponsor conceived and designed the study. Data were collected by the authors and their research teams, analyzed by the sponsor, and interpreted by the authors and sponsor. The sponsor provided for editorial and medical writing support. All authors read and approved the final manuscript before submission. Conflicts of interest O. Hamid, J.R. Stille, J. Polzer, and S. Roberson are employees and stockholders of Eli Lilly and Company. A. Flynt is an employee of PharPoint Research Inc. No potential conflicts of interest were disclosed by the other authors. Acknowledgements The authors wish to acknowledge Sau-Chi Betty Yan (Eli Lilly and Company) for scientific assistance and Jude Richard (INC Research, Austin, TX) for medical writing assistance. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.lungcan.2016.12. 020. References [1] American Cancer Society. Cancer Facts & Figures 2016, American Cancer Society, Atlanta, 2016. [2] J.P. Chute, T. Chen, E. Feigal, R. Simon, B.E. Johnson, Twenty years of phase III trials for patients with extensive-stage small cell lung cancer: perceptible progress, J. Clin. Oncol. 17 (1999) 1794–1801. [3] A. Kurup, N.H. Hanna, Treatment of small cell lung cancer, Crit. Rev. Oncol. Hematol. 52 (2004) 117–126. [4] F. Balkwill, The significance of cancer cell expression of the chemokine receptor CXCR4, Semin. Cancer Biol. 14 (2004) 171–179.

R. Salgia et al. / Lung Cancer 105 (2017) 7–13 [5] S.B. Peng, V. Peek, Y. Zhai, D.C. Paul, Q. Lou, X. Xia, et al., Akt activation, but not extracellular signal-regulated kinase activation, is required for SDF-1␣/CXCR4-mediated migration of epithelioid carcinoma cells, Mol. Cancer Res. 3 (2005) 227–236. [6] M. Kucia, R. Reca, K. Miekus, J. Wanzeck, W. Wojakowski, A. Janowska-Wieczorek, et al., Trafficking of normal stem cells and metastasis of cancer stem cells involve similar mechanisms: pivotal role of the SDF-1-CXCR4 axis, Stem Cells 23 (2005) 879–894. [7] S.B. Peng, X. Zhang, D. Paul, L.M. Kays, W. Gough, J. Stewart, et al., Identification of LY2510924, a novel cyclic peptide CXCR4 antagonist that exhibits antitumor activities in solid tumor and breast cancer metastatic models, Mol. Cancer Ther. 14 (2015) 480–490. [8] M.D. Galsky, N.J. Vogelzang, P. Conkling, E. Raddad, J. Polzer, S. Roberson, et al., A phase I trial of LY2510924, a CXCR4 peptide antagonist, in patients with advanced cancer, Clin. Cancer Res. 20 (2014) 3581–3588. [9] T. Kijima, G. Maulik, P.C. Ma, E.V. Tibaldi, R.E. Turner, B. Rollins, et al., Regulation of cellular proliferation, cytoskeletal function, and signal transduction through CXCR4 and c-Kit in small cell lung cancer cells, Cancer Res. 62 (2002) 6304–6311. [10] J.X. Liang, W. Gao, Y. Liang, X.M. Zhou, Chemokine receptor CXCR4 expression and lung cancer prognosis: a meta-analysis, Int. J. Clin. Exp. Med. 8 (2015) 5163–5174. [11] H. Zhao, L. Guo, H. Zhao, J. Zhao, H. Weng, B. Zhao, CXCR4 over-expression and survival in cancer: a system review and meta-analysis, Oncotarget 6 (2014) 5022–5040. [12] T. Gangadhar, S. Nandi, R. Salgia, The role of chemokine receptor CXCR4 in lung cancer, Cancer Biol. Ther. 9 (2010) 409–416, Epub 2010 Mar 18. [13] A. Schmittel, M. Sebastian, L. Fischer von Weikersthal, P. Martus, T.C. Gauler, C. Kaufmann, et al., Arbeitsgemeinschaft Internistische Onkologie Thoracic Oncology Study Group, A German multicenter, randomized phase III trial comparing irinotecan-carboplatin with etoposide-carboplatin as first-line therapy for extensive-disease small-cell lung cancer, Ann. Oncol. 22 (2011) 1798–1804. [14] H.B. Niell, J.E. Herndon 2nd, A.A. Miller, D.M. Watson, A.B. Sandler, K. Kelly, et al., Randomized phase III intergroup trial of etoposide and cisplatin with or without paclitaxel and granulocyte colony-stimulating factor in patients with extensive-stage small-cell lung cancer: Cancer and Leukemia Group B Trial 9732, J. Clin. Oncol. 23 (2005) 3752–3759. [15] D.H. Johnson, L.H. Einhorn, R. Birch, R. Vollmer, C. Perez, S. Krauss, et al., A randomized comparison of high-dose versus conventional-dose cyclophosphamide, doxorubicin, and vincristine for extensive-stage

[16]

[17]

[18] [19]

[20]

[21]

[22]

[23]

[24] [25]

[26]

13

small-cell lung cancer: a phase III trial of the Southeastern Cancer Study Group, J. Clin. Oncol. 5 (1987) 1731–1738. W.K. Evans, R. Feld, N. Murray, A. Willan, P. Coy, D. Osoba, et al., Superiority of alternating non-cross-resistant chemotherapy in extensive small cell lung cancer: a multicenter, randomized clinical trial by the National Cancer Institute of Canada, Ann. Intern. Med. 107 (1987) 451–458. K. Noda, Y. Nishiwaki, M. Kawahara, S. Negoro, T. Sugiura, A. Yokoyama, et al., Irinotecan plus cisplatin compared with etoposide plus cisplatin for extensive small-cell lung cancer, N. Engl. J. Med. 346 (2002) 85–91. P.K. Wagh, B.E. Peace, S.E. Waltz, Met-related receptor tyrosine kinase Ron in tumor growth and metastasis, Adv. Cancer Res. 100 (2008) 1–33. M. Burger, A. Glodek, T. Hartmann, A. Schmitt-Gräff, L.E. Silberstein, N. Fujii, et al., Functional expression of CXCR4 (CD184) on small-cell lung cancer cells mediates migration, integrin activation, and adhesion to stromal cells, Oncogene 22 (2003) 8093–8101. T. Kijima, G. Maulik, P.C. Ma, E.V. Tibaldi, R.E. Turner, B. Rollins, et al., Regulation of cellular proliferation, cytoskeletal function, and signal transduction through CXCR4 and c-Kit in small cell lung cancer cells, Cancer Res. 62 (2002) 6304–6311. S.P. Treon, C.K. Tripsas, K. Meid, D. Warren, G. Varma, R. Green, et al., Ibrutinib in previously treated Waldenström’s macroglobulinemia, N. Engl. J. Med. 372 (2015) 1430–1440. K.H. Smith, E.Q. Lee, A. Muzikansky, E.R. Gerstner, D.G. Duda, D.A. Reardon, et al., Phase I study of plerixafor and bevacizumab in recurrent high-grade glioma, J. Clin. Oncol. 32 (2014) (suppl: abstr 2031). J.D. Hainsworth, J.R. Mace, J.A. Reeves, E.J. Crane, O. Hamid, J.R. Stille, et al., Randomized phase II study of sunitinib + CXCR4 inhibitor LY2510924 versus sunitinib alone in first-line treatment of patients with metastatic renal cell carcinoma, J. Clin. Oncol. 33 (2015) (suppl: abstr 4547). S.I. Rothschild, O. Gautschi, Crizotinib in the treatment of non-small-cell lung cancer, Clin. Lung Cancer 14 (2013) 473–480. A.K. Shah, V. Sachdew, J. Taylor-Parker, U. Tapan, Y.H. Tohme, Y. Malinkovich, et al., Effect of DNA-PK dependent phosphorylation of topo-I-S10 on its rate of proteasomal degradation and CPT response, J. Clin. Oncol. 33 (Suppl. 3) (2015) (abstr 606). ´ G. Kong, et al., Comprehensive J. George, J.S. Lim, S.J. Jang, Y. Cun, L. Ozretic, genomic profiles of small cell lung cancer, Nature 524 (7563) (2015) 47–53, http://dx.doi.org/10.1038/nature14664.