Results from a biomarker study to accompany a phase II trial of RRx-001 with reintroduced platinum-based chemotherapy in relapsed small cell carcinoma

Abstract

Background:

In a Phase II study RRx-001 was combined with Etoposide platinum (EP) in previously platinum treated SCLC. We correlated expression of the M2 marker, CD206, on HLA-DR^low/- monocytes, a phenotype that correlates with a poor prognosis, with response to RRx-001.

Research design and methods:

Patients received 4 mg RRx-001 once weekly until progression followed by the start of EP (etoposide 100 mg/m² IV on days 1–3 of a 21-day cycle and either cisplatin 80 mg/m² IV on day 1 or carboplatin AUC 5–6 IV on day 1). Treatment continued until progression or intolerable toxicity. Peripheral blood was collected in Cell Preparation Tubes with sodium citrate from 14 patients for exploratory studies during screening and after therapy on Days 1, 8, and 15. Peripheral blood mononuclear cells (PBMCs) were isolated from blood by centrifugation and multiparameter flow cytometric analysis was performed.

Results:

CD206 expression on HLA-DR^low/- monocytes was associated with response to chemotherapy and overall survival.

Conclusion:

During treatment with RRx-001, reduced expression of the protumorigenic M2 marker CD206 on peripheral monocytes positively correlated with increased response and survival.

1. Introduction

Small cell lung cancer (SCLC), which accounts for about 13–15% [1] of all lung cancers and approximately 30,000 new cases per year in the U.S. [2,3], is an aggressive neuroendocrine (NE) tumor derived from bronchial epithelial cells and associated with distinct paraneoplastic syndromes [4], including hypercalcemia, Eaton–Lambert syndrome, and syndrome of inappropriate anti-diuretic hormone. Its rapid doubling time and high growth fraction combined with a propensity to metastasize [5] widely and early in the disease course (most commonly to the brain, liver, or bone) and near universal recurrence after first-line systemic treatment results in a 95% mortality rate [6], making SCLC the most lethal and recalcitrant lung cancer subtype. As a result, the median survival rate is a short 8–13 months [7]. Early diagnosis and effective treatment are two of the key challenges in SCLC given the lack of an effective screening program, the usual presence of metastases at diagnosis and the inevitable development of multidrug resistance after (or during) first-line therapy.

The current standard of first-line care for extensive-stage small cell lung cancer (ES-SCLC) is platinum based on a triplet combination of atezolizumab or durvalumab, carboplatin, and etoposide (alternatively cisplatin and etoposide) with or without concurrent radiation therapy, followed, if applicable, by atezolizumab maintenance and then potentially by topotecan or lurbinectedin [8]. Topotecan binds to the topoisomerase I-DNA complex and prevents the religation of DNA after single-strand cleavage has taken place, which results in cell death [9]. However, topotecan is only minimally effective especially in the platinum refractory setting where the overall response rate and the median overall survival are 6.4% and 4.7 months [10], respectively.

Standard third-line treatments, once lacking, now include nivolumab and pembrolizumab. No treatment is available in fourth line or beyond. Prognosis for patients with relapsed SCLC is poor [11] since relapse is associated with a median overall survival (OS) of <6 months [12], a survival which has scarcely improved over the last 40 years [13].

RRx-001 is a small molecule dual checkpoint inhibitor of SIRP-α and CD-47 [14] with epigenetic modulation [15], vascular normalization [16] and chemoprotective/radioprotective properties [17] that synergize with and reverse resistance to DNA-damaging agents, including radiation [18,19], platinum [20] and topoisomerase inhibitors [21]. Preclinical studies have shown that RRx-001 in combination with platinum and etoposide has a synergistic effect on SCLC xenografts in vivo.

A prerequisite step for RRx-001 activity is thought to be the polarization [22] of tumor-associated macrophages (TAMs) via inhibition of the CD47-SIRP-α axis from a protumorigenic M2-like phenotype to an antitumorigenic M1-like phenotype. In addition to TAMs, RRx-001 may polarize the CD206^high/human leukocyte antigen-antigen D related (HLA-DR)^low phenotype that is associated with immune suppression.

The downstream effects from this M2-to-M1 polarization include:

• epigenetic modulation of DNA methylation and histone acetylation with reactivation of silenced gene transcription since M1 macrophages produce reactive oxygen species (ROS) [23] that inhibit epigenetic enzymes [24]

• anti-angiogenesis [25] and vascular normalization [16] since M2-like TAMs mediate tumor angiogenesis [26]

• reversal of tumor resistance through cancer stem cell (CSC) elimination [27] since M2-like TAMs are known to support and maintain CSCs [28]

• immune stimulation [29,30]since M2-like TAMs mediate immunosuppression [31].

The first-in-man Phase 1 clinical study for single-agent RRx-001 established that it is safe, non-myelosuppressive, and well tolerated, with no observed dose-limiting toxicities and no MTD identified [32]. In addition, Reid et al showed preliminary evidence of anti-tumor activity and durable stable disease (over 2 years) with RRx-001 administered at varying dose levels followed at progression with the off-study reintroduction of standard chemotherapy agents, in which renewed susceptibility or ‘resensitization’ [33–35] to these agents was demonstrated.

In the QUADRUPLE THREAT Phase II study, so-called because four platinum-treated tumor types are eligible, ovarian cancer, high-grade neuroendocrine carcinoma, EGFR+ NSCLC and SCLC, the effect of the sequential combination of RRx-001 and Etoposide Platinum (EP) on the primary end-points of overall survival and overall response rate was evaluated in United States patients with extensive-stage SCLC that was either a) initially non-responsive to standard treatment with EP or b) that relapsed within 90 days of EP treatment or c) that had received at least two prior lines of therapy including EP. The published nonrandomized results [36] in twenty-six patients demonstrated an overall survival (OS) of 8.6 months, a response rate of approximately 27% and a progression-free survival (PFS) of 7.5 months, which is higher-than-expected based on historical data.

Trial endpoints included overall survival (OS), defined as the timeframe between trial enrollment until death due to any cause, and overall response rate (ORR) to platinum plus etoposide (EP), defined as the percentage of patients achieving complete response (CR) or partial response (PR) by RECIST 1.1 criteria.

As a hypothesis-generating clinical trial, exploratory biomarker analysis was performed.

2. Materials and methods

2.1. Patients in phase II clinical trial

In Phase II clinical trial QUADRUPLE THREAT, 26 patients with SCLC were enrolled. Patients were all 18+ years with diagnosed small cell lung carcinoma, previously treated with a platinum-based chemotherapy regimen, with tumor relapse within 6 months after completing their last cycle of first-line chemotherapy and/or having received 2 or more lines of prior therapy.

Patients were administered 4 mg RRx-001 once weekly until progression at which point EP was started. The EP chemotherapy consisted of etoposide 100 mg/m² IV on days 1–3 of a 21-day cycle and either cisplatin 80 mg/m² IV on day 1 or carboplatin AUC 5–6 IV on day 1. Study treatment with RRx-001 and chemotherapy continued until progression or intolerable toxicity.

Peripheral blood was collected in Cell Preparation Tubes with sodium citrate from 14 patients for exploratory studies during screening and after therapy on Days 1, 8, and 15. At the discretion of the clinical investigators, tissue biopsy was also obtained during screening and after therapy on Days 1, 8, and 15. Samples for this biomarker study were obtained at the discretion of the clinical investigators, and therefore samples were not obtained for all 26 patients in the SCLC cohort of the clinical trial.

2.2. Circulating biomarker evaluations

Peripheral blood was collected in Cell Preparation Tubes with sodium citrate (BD Vacutainer CPT Tubes; BD Biosciences, San Jose, CA) from 13 patients for exploratory studies during screening and after therapy in cycle 1 (Days 1, 8, and 15). Human mononuclear cells were isolated from blood by centrifugation. Briefly, the circulating mononuclear cells were collected, fixed in 4% paraformaldehyde at 4°C for 0.5–2 h, blocked with 5% bovine serum albumin for 20 min at room temperature, and stained with HLA-DR, CD206, or signal regulatory protein alpha (SIRP) antibodies at 4°C for 20 min. The stained cells were then analyzed by flow cytometry with FlowJo software.

2.3. Immunohistochemical evaluations

Formalin-fixed, paraffin-embedded tissues were sectioned at 5 microns for immunohistochemical staining. Paraffin was removed in xylene solution for 5 min × 3, and tissue was rehydrated by placement for 5 min each in a graded series of progressively diluted ethanol. Antigen retrieval was performed by placement of slides in 10 mM sodium citrate buffer, pH 6.0, at boiling temperature for 10 min. Samples were then cooled on the benchtop for ~30 min. Sections were washed 3 × with H₂O 5 min at room temperature, followed by placement in 3% H₂O₂ for 10 min to block endogenous peroxidase activity. Sections were rinsed in H₂O for 5 min at room temperature followed by placement in 1 × phosphate-buffered saline (PBS). Sections were blocked in 1 × animal serum in blocking buffer (Cell Signaling Technology, Danvers, MA) for 30 min at room temperature. Sections were stained manually with anti-CD68 and anti-CD163. The CD163 and CD68 staining was scored as the infiltration density of CD163+ and CD68+ cells with a monocyte/macrophage morphology, ranging from 0 (absent) up to 3 (dense). These categories were dichotomized into absent/sparse (0–2) or dense (3) macrophage infiltration. The immunohistochemical staining was evaluated by a board-certified pathologist in a blinded fashion.

2.4. Statistical analyses

To understand whether changes in CD206 expression on peripheral HLA DR^low monocytes may influence response to therapy, patients were dichotomized into two groups based on whether its expression was increased or decreased (relative to baseline) on treatment. Spearman correlations were used to quantify the relationship between the change in CD206 expression on HLA DR^low monocytes at baseline, Day 8, and Day 15 of treatment with RRx-001 and outcome. Spearman, Pearson, and Kendall correlation estimates were consistent.

Kaplan–Meier survival analysis was performed for correlation between OS and the change in CD206 expression on HLA DR^low monocytes between baseline, Day 8, and Day 15.

All data were analyzed using SAS.

2.5. Ethical approval

The study protocol and consent were approved by the relevant Institutional Review Boards of each institution participating in the clinical trial.

3. Results

3.1. Analysis of biopsies for tumor-associated macrophage expression

Biopsy tissue was available from eight patients, four responders and four non-responders, for testing of TAM expression by immunohistochemistry (IHC). TAM expression and density were assessed by single immunohistochemical staining of CD206, which is present in protumoral M2 macrophages. The patients were divided into two groups according to the mean number of TAMs: low and high expression groups. In the Kaplan–Meier survival curves, the outcome of patients in the high CD206⁺ expression group was significantly improved over those in the low expression group.

3.2. Circulating biomarker analysis

In an exploratory analysis of the relationship between biomarkers and efficacy, a strong correlation was present between relative change in CD206 on HLA DR^low monocytes (between baseline, Day 8, and Day 15) and the objective response rate (Pearson’s coefficient 61.4%). Figure 1 shows the relative change of CD206 levels on HLA DR^low monocytes compared to baseline during RRx-001 (and EP) treatment. A decrease in the level of CD206 that decreased during treatment was associated with improved OS (HR 0.785) and, conversely, increase of CD206 during treatment was associated with poorer OS (HR 1.27), which is visualized with a Kaplan–Meier curve in Figure 2. The hazard ratio was estimated using the Cox proportional hazard model.

Figure 1.

Relative change of CD206 levels on HLA DR^low monocytes compared to baseline during RRx-001 treatment. Each line indicates values for single patients. The fan-shaped pattern of change relative to baseline suggests a dichotomy between responders and non-responders based on CD206 deviation, downward or upward, respectively, from baseline. C = cycle; D = day.

Figure 2.

OS Kaplan–Meier Curves by monocyte CD206 decrease (group 1) vs. monocyte CD206 increase (group 2).

3.3. Immunohistochemistry

From a blinded analysis by two pathologists, staining of the only four available pre- and post-treatment tumor biopsies with the pan-macrophage marker CD68 increased in patients responsive to RRx-001 (Figure 3). The one biopsied patient that was non-responsive to RRx-001 displayed no change in CD68 staining. Patients responsive to RRx-001 survived for greater than 12 months; the one non-responsive patient died within 4 months.

Figure 3.

CD68 positive macrophages increase in tumor biopsies from patients responsive to RRx-001.

3.4. Limitations of this study

Within the Phase II clinical trial, the patients received identical treatment regimens according to the clinical trial protocol. This report describes the treatment regimen of the patients for the duration of Phase II clinical trial. This report does not publish potential past treatments that the patients may have received in years prior to enrolling in the Phase II QUADRUPLE THREAT clinical trial. However, each patient received no treatment for a defined period prior to trial enrollment.

Biopsies and samples were collected at the discretion of the clinical investigators. For each patient, pre- and post-biopsies were taken from the same locations when it was feasible. Macrophage density can vary intratumorally and this effect was accounted for and potentially offset by the collection of multiple biopsies.

When interpreting the results of this study, the reader should take into consideration the limited sample sizes. These experiments are based on patient samples from a Phase II clinical trial with limited patient cohort size and do not constitute a ‘statistically significant’ study population. While larger patient populations are unavailable at the time of publishing, the authors believe that there is value in publishing these results, for the education and insight of researchers and clinicians alike when evaluating novel treatment strategies involving RRx-001. Follow-up biomarker studies could be performed in the future in conjunction with the ongoing Phase 3 REPLATINUM clinical trial featuring RRx-001 + platinum doublet treatment in late line small cell carcinoma in which a larger patient population is included in the study.

4. Discussion

The presence of TAMs intratumorally or in the tumor microenvironment has been correlated with poor prognosis in several human studies due to the facilitation of tumor growth and disease progression [37,38]. In contrast, nonclinical studies have shown that the more extensive the macrophage infiltration in tumors, the better the response to RRx-001 [39]. In the QUADRUPLE THREAT trial, although the number of samples was limited, due to the reluctance of clinical investigators to perform biopsies, an association was observed between dynamic TAM density and response to RRx-001. In three patients with durable disease control for over 1 year the level of TAM infiltration increased. However, in one patient that did not respond to RRx-001 or EP the level of TAM infiltration, relatively low to begin with, remained constant. While the limited number of biopsy samples as well as the lack of a comparator makes it impossible to establish a true association between dynamic TAM density and benefit of RRx-001 in combination with EP a suggestive signal is present.

In addition, there was a strong correlation between expression of CD206, a phenotypic marker for M2 monocytes/macrophages, on HLA DR^−/low monocytes and outcome on trial. HLA DR^−/low monocytes are myeloid suppressor cells associated with decreased time to progression in several cancers including non-small cell and small cell lung cancer [40,41]. In general, improved overall survival was seen in QUADRUPLE THREAT SCLC patients where CD206 levels decreased on HLA DR^−/low monocytes, which suggests that RRx-001 skews their polarization from a protumor M2-like toward an antitumor M1-like phenotype and may reduce the development of metastases.

Moreover, in an in vitro study, RRx-001 was shown to decrease the surface expression of SIRPα on nonclassical monocytes, the main subpopulation of blood monocytes involved in antitumor response [42]. The downregulation of SIRPα on non-classical monocytes would be expected to enhance the efficiency of engulfment of disseminated tumor cells, which is suggestive because only one of the evaluable patients that progressed on trial developed new metastatic lesions.

Monocyte chemoattractant protein-1 (MCP-1), also referred to as chemokine (C-C motif) ligand 2 (CCL2), has been identified as an important chemokine regulating the migration and infiltration of monocytes/macrophages. Based on these correlative data, it is tempting to speculate that serum or urine levels of MCP-1 (monocyte chemoattractant protein-1) [43], a proinflammatory cytokine that actively recruits monocytes to tumors, may serve as a biomarker predicting a priori, rather than a posteriori whether individual patients will respond to treatment with RRx-001.

5. Conclusions

During treatment with RRx-001, reduced expression of the protumorigenic M2 marker CD206 on peripheral monocytes positively correlated with increased response and survival, suggestive of a systemic increase in the ratio of M1/M2 macrophages that may, in turn, attenuate the metastatic process. Given the limited sample size of this study, further studies are warranted to confirm the validity of these results. A Phase III trial with RRx-001 in SCLC called REPLATINUM is underway in which the prediction potential of TAM density on response to treatment will be further evaluated.