Deep and durable response to nivolumab and temozolomide in small cell lung cancer associated with an early decrease in myeloid-derived suppressor cells.

Structured Abstract:

Background:

Immune checkpoint inhibitors are now an approved treatment for patients with extensive-stage small cell lung cancer (SCLC), an aggressive and incurable malignancy. However, the median survival for patients remains around 1 year, and treatment in the second line and beyond is associated with a low likelihood of response. There currently are no data regarding the optimal treatment of patients who progress on upfront chemo-immunotherapy, and biomarkers are needed to aid patient selection. Obtaining sufficient tissue for advanced molecular testing is a challenge in SCLC due to often limited or crushed tissue specimens. Reliable blood based biomarkers may augment tissue based testing and allow for repeated assessments that is often not possible with tumor tissue, and changes in the peripheral blood may provide information regarding the tumor microenvironment. Myeloid-derived suppressor cells (MDSC) exert immunosuppressive function and have been investigated as a potential barrier to response to immune checkpoint inhibitor (ICI) therapy. Here we demonstrate that MDSC level and function can be readily assessed in the peripheral blood at multiple time points during treatment with combination immunotherapy and temozolomide.

Results:

This report represents the first case of a patient with refractory SCLC treated with combination nivolumab and temozolomide as part of a clinical trial (NCT03728361), who sustained a deep and durable clinical response that was accompanied by an early decrease in MDSC and improved T cell function (increased CD8⁺ and CD4⁺ T cell proliferation). We review the literature regarding use of ICI in SCLC and the evidence supporting MDSC as a possible target to enhance the activity of immunotherapy, and emphasize the importance of assessing immune cell subsets as correlative studies in clinical trials.

Conclusion:

An assessment of MDSC level and function during treatment, as well as other immune cell subsets, should be included in prospective studies to further evaluate these assays as possible blood-based biomarkers.

Introduction:

Small cell lung cancer (SCLC) is an aggressive malignancy with a median survival of just over one year despite the recent introduction of combination chemotherapy-immunotherapy.¹ Responses in the second line setting are low, especially in patients whose tumors are refractory to chemotherapy or in cases where cancer recurs within three months of therapy.² Temozolomide and nivolumab as single agents are associated with modest clinical responses in the second line setting and beyond.^3–5 The immunomodulatory impact of these treatments is poorly understood, and further knowledge regarding how combination modalities affect antitumor immunity are needed to guide rational combination therapies. Because SCLC is usually diagnosed at an advanced stage, tumor biopsies are often of cytology specimens of metastatic sites and may be limited by modality (i.e fine needle aspirate) or crush artifact which is common in SCLC.⁶ Blood based biomarkers may overcome these limitations and provide valuable information regarding likelihood of response to therapy.⁷ In an ongoing clinical trial of combination temozolomide and nivolumab in recurrent SCLC (NCT03728361), we are conducting a comprehensive assessment of circulating immune cells to determine whether baseline levels or changes in immune cell subsets may predict response to therapy. Here we present the initial report of a patient with recurrent, platinum-refractory SCLC treated with temozolomide and nivolumab who sustained a dramatic and durable partial response that was accompanied by a decrease in myeloid-derived suppressor cells (MDSC) and improved T cell proliferation. To our knowledge, this is the first report of changes in MDSC during treatment with chemo- immunotherapy for SCLC, and point to the importance of including comprehensive immunophenotyping in ongoing clinical trials to aid in patient selection and prediction of clinical responsiveness.

Discussion:

Case Presentation:

A 68 year-old man with a 45-pack year history of smoking presented with hemoptysis leading to acute respiratory failure requiring invasive mechanical ventilation. Bronchoscopy revealed a bleeding left sided endobronchial tumor. Cryoablation was utilized to control bleeding, and biopsy of the tumor along with endobronchial ultrasound guided transbronchial needle aspiration of a lower paratracheal (4R) lymph node revealed histology consistent with SCLC. Immunostains were positive for synaptophysin, chromogranin, CD56, and Ki-67 was 70% by manual estimation. The patient was successfully weaned from mechanical ventilation. Computed tomography (CT) of the chest, abdomen and pelvis showed a left hilar mass consistent with the patient’s primary lung cancer as well as enlarged mediastinal lymph nodes and extensive metastatic liver lesions (Figure 1). The patient was treated with platinum-based chemotherapy and had an excellent partial response after two cycles (Figure 1). Of note, his treatment course started prior to the approval of first line atezolizumab. He required a dose reduction from cycle 2 onwards for thrombocytopenia (carboplatin AUC 5, etoposide 75 mg/m²) but otherwise tolerated therapy well. Disease progression was observed in the lung and liver on imaging performed two months after completion of treatment in a pattern consistent with the initial presentation (Figure 2A). Additionally, new brain metastases were also detected on restaging MRI brain. The patient received whole brain radiotherapy and was consented to an investigator-initiated phase 2 clinical trial of temozolomide and nivolumab (NCT03728361, Sponsor: Ohio State University).⁸

Figure 1. Baseline and response to first line treatment.

Computed tomography images of chest, abdomen and pelvis at the time of diagnosis (Figures 1A and 1C) and after 2 cycles of first-line treatment with standard of care chemotherapy showing response in lung (indicated by asterisk) and liver lesions (Figures 1B and 1D).

Figure 2. Clinical response to nivolumab and temozolomide is accompanied by decreased total MDSC and increased T cell proliferation.

A-C: Representative images showing complete resolution of a left lower lobe lung mass (*) after 8 weeks of study therapy with ICI and chemotherapy (B) that was sustained at 24 weeks (C). Circulating levels of total MDSC (CD11b⁺, CD33⁺, HLA-DR^lo/−), monocytic (M-MDSC; CD14⁺) and granulocytic (PMN-MDSC; CD66b⁺) MDSC were measured by CyTOF at screening and at cycle 1, day 15 of nivolumab and temozolomide treatment and summarized in (D). Patient PBMCs were labeled with carboxyfluorescein succinimidyl ester (CFSE) and activated with anti-CD3/CD28 beads. Following 72 hours, cells were collected and stained with anti-CD8 (PE-Cy7) and anti-CD4 (APC) antibodies and proliferation was assessed by flow cytometry. Histograms and bar graphs of fold change quantification between screening and cycle 1, day 15 (C1D15) are shown for (E) CD8⁺ T cell proliferation and (F) CD4⁺ T cell proliferation.

Immunohistochemistry of his tumor biopsy for PD-L1 was performed on the original biopsy, and tumor proportion score (TPS) was found to be 0% using the 22C3 antibody assay. The patient was treated with nivolumab (480 mg intravenous on day 1 of 28-day cycle) and temozolomide (150 mg/m² orally days 1-5 of 28-day cycle). He had a complete resolution of his lung lesion and confirmed partial response in liver lesions (overall 82% decrease in target lesions by RECIST v1.1) after 2 cycles of treatment which is ongoing after 40 weeks of study therapy (18 months since initial diagnosis of extensive stage SCLC) (Figure 2B–C). The patient has tolerated therapy well without serious adverse events, had improvement in breathing and fatigue, and returned to working part-time. Treatment-related adverse events included grade 1 thrombocytopenia, transient grade 3 lymphopenia (ALC 0.45 K/uL), grade 1 rash later diagnosed as worsening of baseline rosacea, and grade 2 weight loss and pre-syncope attributed to temozolomide which both resolved without dose modification or delay. As part of the study, peripheral blood was collected at baseline and after 2 weeks of treatment. Mass cytometry (CyTOF) analysis revealed an early decrease (19.8% to 16.7%) in total and PMN-MDSC (Figure 2D) an increase in CD8⁺ central memory (CM) and effector memory (EM) T cells (Supplemental Figures 1 and 2) as well as an increase in CD8⁺ and CD4⁺ T cell proliferation during therapy (Figure 2E–F). CD4⁺ and CD8⁺ T cells were phenotyped into the following subsets based on the expression of CD45RA and CCR7, naive (CD45RA⁺ CCR7⁺), central memory (CD45RA⁻ CCR7⁺), effector memory (CD45RA⁻ CCR7⁻), and terminal effector T (CD45RA⁺ CCR7⁻) T cells (Supplemental Table 2).⁹

Discussion:

The prognosis of patients diagnosed with extensive-stage SCLC remains poor despite the introduction of immune checkpoint inhibitor (ICI) therapy in the first line setting, with a median survival of around 1 year.¹ A number of second line options are available, yet most are associated with modest clinical benefit. Prior studies have demonstrated that temozolomide and nivolumab each have activity as single agents in the second line setting and beyond.^{3, 5} However, the likelihood of response is lower in patients with platinum-refractory disease than for patients with platinum-sensitive disease.² Reliable predictors for benefit from ICI in SCLC remain unclear, as PD-L1 expression in SCLC has not shown the same predictive ability as in non-small cell lung cancer.³ Overall, the rate of PD-L1 expression within tumor cells is low in SCLC,¹⁰ although has been reported to be higher in tumor infiltrating cells.¹¹ Assessment of tumor-based biomarkers in clinic trials has been limited by the availability of sufficient specimens; even in prospective clinical trials at high volume centers, availability of suitable tumor specimens range from 30-50%, significantly limiting the ability to evaluate for predictive biomarkers.^{4, 5} Blood based biomarkers offer several advantages over tumor biopsies including avoiding invasive procedures and allowing for repeated assessments during treatment to assess response. Several blood based biomarkers have been explored in SCLC including cell free DNA, circulating tumor DNA, and circulating tumor cells.^{7, 12, 13}. Recently, the importance of understanding the immune response to neoplastic therapy to predict efficacy of immunotherapies has been demonstrated in pilot studies in patients with non-small cell lung cancer.^{14, 15} However, an understanding of immune cell changes in patients with SCLC receiving combination therapy is lacking.

MDSC have been shown to exert immunosuppressive effects in many cancers, and levels of circulating MDSC have been associated with a worse prognosis in patients with SCLC receiving platinum-based chemotherapy.¹⁶ Notably, levels of MDSC have been associated with patient survival in ipilimumab-refractory melanoma and platinum-resistant urothelial carcinoma patients treated with nivolumab.^{17, 18} Previous studies have reported a decrease in MDSC numbers with chemotherapeutic agents however, the impact of chemo-immunotherapy on MDSC in patients with SCLC is unknown, nor is it clear that changes in MDSC levels will be able to predict response to treatment.^19–21 Lastly, whether targeting MDSC may improve outcomes of patients treated with ICI or overcome resistance to ICI is unknown. In an ongoing clinical trial, we observed a decrease in circulating total MDSC, an increase in CD4⁺ and CD8⁺ T cell proliferation, and an increase in central memory and effector memory T cells which was accompanied by a robust and durable response to treatment with combination nivolumab and temozolomide in a patient with platinum-refractory SCLC. Although an increase in both CD4⁺ and CD8⁺ central and effector memory T cells was observed, the increase was greater in CD8⁺ T cells (Supplemental Figure 1). Further studies are needed to determine the significance of this finding, however, increases in memory T cells have been correlated with clinical response in patients with melanoma and NSCLC treated with PD-1 inhibition.^22–24

Immunotherapy with checkpoint inhibitors has improved outcomes for patients with SCLC. In a phase I/II trial (CheckMate 032, NCT01928394) which included 98 patients with extensive stage SCLC who received nivolumab monotherapy for SCLC, objective responses were seen in 10% (n=10) of patients whose disease progressed after at least 1 prior therapy.³ In this study, responses were observed regardless of programmed death ligand-1 (PD-L1) expression or platinum sensitivity/resistance. These encouraging data led to a phase III trial, CheckMate 331 (NCT02481830), which was designed to evaluate nivolumab monotherapy versus chemotherapy in patients with relapsed SCLC. Unfortunately, this study did not meet its primary endpoint of improvement in overall survival compared to standard of care chemotherapy (topotecan or amrubicin).²⁵ In the final analysis of Checkmate 032, a total of 109 patients were treated with nivolumab in at least third line setting, with an objective response rate of 11.9% (95% CI: 6.5–19.5) and a median duration of response of 17.9 months (95% CI: 7.9–42.0).²⁶ Based on these data, nivolumab is now approved as a third line treatment option. In a pooled analysis of two trials evaluating 83 patients with SCLC in the second or third line setting treated with pembrolizumab, responses were observed in 19.3% (n=16) of patients with both PD-L1 positive and negative tumors.²⁷ The study reported a median PFS of 2.0 months and median OS of 7.7 months, leading to the approval of pembrolizumab in the third line setting.

Recently, the addition of the anti-PD-L1 antibody atezolizumab to carboplatin and etoposide chemotherapy demonstrated an improvement in OS in the first-line setting (median OS of 12.3 months vs. 10.3 months for chemotherapy alone [95% confidence interval: 0.54–0.91; P=0.007]), leading to the approval of this regimen by the U.S. Food and Drug Administration for first-line treatment of extensive stage SCLC.^{28, 29} Subsequently, a randomized phase III study of durvalumab in combination with platinum chemotherapy reported similar findings.³⁰ Although the addition of an immune checkpoint inhibitor to chemotherapy prolonged survival, the limited median survival benefit of two months highlights the need for novel treatment approaches. Treatment of SCLC beyond the first line is associated with low response rates despite decades of clinical trials.^31–34 Recently lurbinectedin was approved as second line therapy after results from a phase 2 basket study including 105 patients with SCLC demonstrated a response rate of 35% (95% CI 26.2–45.2) in a single arm, open-label study of patients with SCLC who had progressed after first line chemotherapy.³⁵ The response rate was 22% and median duration of response was 4.7 months for patients with platinum-resistant disease. However the study did not include patients with brain metastases and only 3 patients received chemo-immunotherapy as first line treatment, so further studies are needed to evaluate efficacy in these patient populations. There remain no available data regarding the activity of single agent chemotherapy from dedicated clinical trials in patients who have progressed on first line combination chemotherapy and immunotherapy. Temozolomide has been studied in prospective phase 2 trials with response rates ranging from 12% to 23% in two different dosing strategies.^{4, 5} The 5-day dosing schedule was associated with lower incidence of prolonged pancytopenia and is therefore most commonly utilized in this setting.⁵

Levels of myeloid-derived suppressor cells (MDSC) have been shown to be a poor prognostic indicator in many cancers including SCLC.¹⁶ Levels of circulating MDSC are higher in patients with larger tumors or with higher stage of cancer in several tumors.^{36, 37} MDSC have also been associated with decreased responsiveness to ICI through the inhibition of specific anti-tumor immunity and promotion of an immunosuppressive tumor microenvironment. Pre-clinical studies have shown that modulation of MDSC through a variety of mechanisms can enhance T-cell responsiveness to immunotherapy.^38–40 MDSC are divided into two major subsets, monocytic (M) and granulocytic or polymorphnuclear (PMN)-MDSC, based on their phenotypic and morphologic features.^41–43 These MDSC subsets are suggested to have differential effects in their ability to suppress the immune system, yet further investigation is needed to clearly understand the role played by MDSC in promoting progression.⁴⁴ Temozolomide has been demonstrated to have immunomodulatory effects on lymphoid cells in melanoma patients.^{45, 46} Different temozolomide dosing strategies have been shown to exert varying effects on MDSC and regulatory T cells in a murine model of glioblastoma.⁴⁷ The utility of targeting MDSC to affect anti-tumor immunity was evaluated in a randomized phase 2 trial evaluating the role of a dendritic cell vaccine in patients with SCLC which reported that modulation of MDSC through all-trans retinoic acid led a two-fold decrease in MDSC compared to vaccination alone.⁴⁸ However, the study was not powered to detect any differences in clinical outcomes and did not meet its primary endpoint. Changes in MDSC in patients with SCLC occurring during immunotherapy with checkpoint inhibitor therapy have not been studied. This research provides important preliminary evidence for the clinical activity of temozolomide and PD-1 therapy in refractory SCLC and points to the importance of including comprehensive assessment of immune cell changes as a biomarker of response to treatment.

Although promising, these findings should be interpreted with caution. Possible synergy between nivolumab and temozolomide is impossible to demonstrate in this case report since responses are seen with both agents given as monotherapy. As there is a lack of data on changes of MDSC during treatment with ICI, the impact of quantitative changes in MDSC levels remains unclear, pointing to the importance of including these analyses in prospective studies. The optimal treatment for patients who progress on first-line chemo-immunotherapy, as well as how MDSC are involved and impacted remain to be studied. Lastly, although first line chemo-immunotherapy is standard of care, there remain no data on changes in immune subsets in these patients and clinical relevance, thus the translational studies presented here are hypothesis-generating for future research. Accrual to clinical trial NCT03728361 is ongoing and is now accruing only patients who progress on first-line chemo-immunotherapy to evaluate this regimen in the current treatment landscape.

Conclusion:

This is the first report of treatment with combination temozolomide and nivolumab in progressive refractory extensive stage small cell lung cancer. This is also the first research to demonstrate that changes in MDSC can be detected during treatment with chemo-immunotherapy and points to the importance of including these correlative studies in prospective clinical trials.

Highlights:

• -Platinum resistant small cell lung cancer is characterized by low response rates and short survival.

• -This case demonstrates sustained response to nivolumab and temozolomide as part of a clinical trial in platinum resistance small cell lung cancer.

• -The clinical response was accompanied by a decrease in immunosuppressive myeloid derived suppressor cells which may be early biomarker of response to treatment.

Clinical Practice Points:

Small cell lung cancer is an aggressive malignancy, and treatment in the second line and beyond is associated with low response rates. Platinum resistant small cell lung cancer, where disease progression occurs within 3 months of platinum doublet, indicates a particularly poor prognosis. Here we present the first report of combination nivolumab and temozolomide given as part of clinical trial NCT03728361 in platinum-resistant small cell lung cancer, which led to durable response that is ongoing at 40 weeks of therapy. As part of the clinical trial, peripheral blood was collected for comprehensive immune cell evaluation by mass cytometry (CyTOF), which allows for assessment of multiple targets per cell. In this case, CyTOF analysis revealed a decrease in immunosuppressive myeloid-derived suppressor cells and increase in T cells. Accrual to clinical trial NCT03728361 is ongoing in hopes of addressing both the clinical efficacy of this combination as well as impact on immune cell subsets which may serve as a biomarker of response.