Phase 1 trial of CV301 in combination with anti-PD-1 therapy in nonsquamous non-small cell lung cancer

Abstract

CV301, a poxviral-based vaccine, has been evaluated in a phase 1 clinical trial (NCT02840994) and shown to be safe and immunologically active (phase 1a). Preclinical data support a combination of CV301 with programmed death-1 inhibitors, which has been evaluated in the phase 1b part of this trial and is reported here. Patients with advanced nonsquamous non-small cell lung cancer (NSCLC) without actionable genomic alterations received two priming doses of modified vaccinia Ankara-BN-CV301 (MVA) 4 weeks apart, followed by boosting doses of fowlpox-CV301 (FPV) at increasing time intervals for a maximum of 17 doses in combination with nivolumab for cohort 1 (C1) and 15 doses in combination with pembrolizumab for cohort 2 (C2). The primary objective was evaluation of safety and tolerability. Between October 2017 and September 14, 2018, patients were enrolled (C1: 4; median age: 64 years). Mean treatment duration was 332 days in C1 and 289 days in C2. CTCAE ≥grade 3 adverse events (AEs) were observed in four (100%) patients in C1 and three (37.5%) patients in C2. There was one death on trial. Immune-related AEs (irAEs) fulfilling criteria for a dose-limiting toxicity included 1 case of pneumonitis. Among 11 evaluable patients, 1 (9%) had a complete response, 1 (9%) had a partial response and 9 (82%) had stable disease. We conclude that CV301 administered with PD-1 inhibitors is safe and clinically active in patients with advanced NSCLC. The frequency or severity of AEs is not increased, including irAEs for each component of the combination.

1 |. INTRODUCTION

The clinical development of immune checkpoint inhibitors (ICIs) has rapidly altered the landscape of treatment options for patients with non-small cell lung cancer (NSCLC). ICIs, when used in combination with chemotherapy for frontline treatment of advanced NSCLC or as monotherapy for tumors with high expression of programmed-death ligand-1 (PD-L1), have resulted in substantial improvements in objective response rate (ORR) and overall survival (OS).^1,2 However, the benefits of ICI therapy are limited by intrinsic and acquired resistance to treatment.^3,4 To overcome these challenges, combinations of ICIs with various interventions, including other immunotherapeutic modalities, are under investigation.^5,6

Historically, the use of vaccines to elicit an antitumor immune response has been associated with limited benefits.⁷ In advanced NSCLC, CIMAvax-EGF, a therapeutic vaccine targeting the interaction between epidermal growth factor (EGF) and its receptor (EGFR) was associated with an improvement in OS in the per-protocol population when used as maintenance therapy after first-line chemotherapy.⁸ However, belagenpumatucel-L, an allogenic tumor cell vaccine, did not improve survival in the same patient population.⁹ Previous studies have demonstrated the advantages of using cancer vaccines with ICIs to improve antigen presentation and tumor antigen-specific T-cell number and function.⁴

CV301, a highly attenuated, purified poxvirus-based vaccine consisting of priming doses of recombinant modified vaccinia Ankara (MVA-BN-CV301) and boost doses of recombinant fowlpox vaccine (FPV-CV301), contains transgenes encoding two human tumor-associated antigens (TAAs), mucin-1 (MUC-1) and carcinoembryonic antigen (CEA), along with three costimulatory molecules: B7.1, intercellular adhesion molecule-1 (ICAM-1) and leukocyte function-associated antigen-3 (LFA-3).¹⁰ In the phase 1a portion of the clinical trial described here, the vaccine was administered using a prime-boost strategy and was shown to be safe and capable of generating antigen-specific T cells to MUC-1 and CEA.¹⁰ Clinical activities were also observed, especially in patients with KRAS-mutated colorectal cancer. Immune responses to vaccinia do not block infection and immunization with fowlpox-based vectors. Hence, vaccinia-primed immune responses can be boosted with fowlpox vectors.

MUC-1 is a heterodimeric glycoprotein that is overexpressed in greater than 80% of NSCLCs, facilitates the development of an immunosuppressive tumor microenvironment by upregulation of PD-L1 and suppression of interferon-γ (IFN-γ), and is associated with poorer survival in NSCLC patients.^11,12 MUC-1-targeting vaccines have been shown to elicit antigen-specific T-cell responses and improved clinical outcomes in advanced NSCLC.^13,14 CEA is a TAA belonging to the immunoglobulin supergene family that is overexpressed in 70% of NSCLCs and can cause cellular transformation and inhibit cell death.¹⁵ Cancer vaccines targeting CEA have been extensively evaluated and shown to induce humoral and cellular antitumor immune responses.¹⁵

The ability of CV301 to generate immune responses to MUC-1 and CEA and the aberrant overexpression of these glycoproteins in NSCLC provide a rationale for evaluation of the vaccine in patients with advanced NSCLC. Further, concurrent targeting of the inhibitory immune checkpoint, PD-1, could potentially augment the T cell-mediated immune response generated by the vaccine by blocking the inhibitory signal associated with PD-1. The phase 1b trial reported here was conducted to test this hypothesis by evaluating CV301 in combination with nivolumab or pembrolizumab in patients with advanced nonsquamous NSCLC.

2 |. PATIENTS AND METHODS

This open-label phase 1a/1b parallel design trial was conducted in the United States between 2016 and 2020 (NCT02840994). This article focuses on the phase 1b portion of the trial. The phase 1a portion of the trial has been reported previously.¹⁰

2.1 |. Trial procedures

The primary objective for phase 1b was to evaluate the safety and tolerability of CV301 in combination with anti-PD-1 therapy, and the primary endpoint was the occurrence of dose-limiting toxicities (DLTs). Exploratory efficacy endpoints included assessment of ORR, OS, progression-free survival (PFS) and duration of response (DOR).

Enrollment of 12 NSCLC patients meeting eligibility criteria was planned, with all patients to be treated with nivolumab + CV301. During the course of the trial, pembrolizumab was approved for first-line treatment of metastatic NSCLC either with or without pemetrexed and carboplatin based on the results of the KEYNOTE-189 and KEYNOTE-024 studies.^16,17 Hence, the protocol was updated to include an arm for pembrolizumab + CV301, with patients split into two cohorts depending on their treatment history and current regimen. Cohort 1 (C1), nivolumab + CV301, included patients with disease progression on or after one prior line of platinum-based chemotherapy. Cohort 2 (C2), pembrolizumab + CV301, included patients who were on pembrolizumab as first-line therapy for NSCLC for at least 11 weeks and assessed by Response Evaluation Criteria in Solid Tumors (RECIST) to have an ongoing response or disease stability at 12 weeks of treatment. Doses of nivolumab and pembrolizumab used in our trial were 240 mg intravenously (IV) over 30 minutes every 2 weeks and 200 mg IV over 30 minutes given every 3 weeks, respectively. Dosing for CV301 was scheduled to correspond to nivolumab or pembrolizumab dosing. C1 received MVA-BN-CV301 with a nominal virus titer of 4 × 10⁸ infectious units (Inf.U)/0.5 mL on weeks 1 and 5, followed by FPV-CV301 with a nominal virus titer of 1 × 10⁹ Inf.U/0.5 mL every 2 weeks through week 15, every 4 weeks through week 51 and every 13 weeks through end of treatment (EOT) during week 104. C2 received MVA-BN-CV301 4 × 10⁸ infectious units Inf. U/0.5 mL on weeks 1 and 4, followed by FPV-CV301 1 × 10⁹ Inf. U/0.5 mL every 3 weeks through week 22, every 6 weeks through week 52 and every 12 weeks through EOT during week 100. MVA-BN-CV301 was administered as four injections (one in each arm and leg), and FPV-CV301 was administered as a single injection, preferably in the nondominant arm. Dosing of CV301 was based on the recommended phase 1a dose and all doses were administered subcutaneously with a 24- or 25-gauge needle according to standard clinical practice.

The treatment phase for all patients ended after every enrolled patient had completed the week 51/52 visit.

Men and nonpregnant women at least 18 years of age were eligible if they had histologically confirmed, metastatic or unresectable locally advanced nonsquamous NSCLC. Key exclusion criteria included presence of EGFR-sensitizing mutations, or anaplastic lymphoma kinase (ALK) translocations or ROS-1 fusions, concurrent investigational agents, chemotherapy, or other anti-PD-1/L1 therapy, an uncontrolled recent infection or intercurrent illness, a history of or actively ongoing serious medical conditions including autoimmune disease or malignancy, other malignancies within the last 5 years, presence of untreated brain metastases, allergy to vaccine or ingredients, known immunodeficiency, concurrent use of systemic corticosteroids, clinically significant heart disease or cerebrovascular accident within 1 year and/or other medical or psychological impediments which would impede compliance with the protocol.

All trial-related procedures were performed in accordance with the provisions of the Declaration of Helsinki (2013). Informed consent was obtained from all patients for participation in the trial. Ongoing safety oversight was conducted by local IRBs and a safety monitoring committee. Serious adverse events (SAEs) were reported to the United States Food and Drug Administration for review, per established guidelines.

2.2 |. Safety assessment

Safety assessments included adverse events (AEs), DLTs, SAEs and AEs of special interest (AESIs). Since the majority of AEs observed in the phase 1a portion of the trial occurred during the priming dose of the vaccine (MVA-BN-CV301) and decreased over time,¹⁰ and most immune-related adverse events (irAEs) associated with ICIs also occur within 15 weeks of initiation of treatment,¹⁸ AEs reported by the participant or observed by the investigator in the phase 1b portion of the trial were recorded from the time informed consent was obtained through 30 days after the last dose of CV301 was administered, or through 100 days post-final treatment for treatment-related SAEs or AESIs. AEs were graded according to the National Cancer Institute Common Toxicity Criteria for Adverse Events (NCI-CTCAE) version 4.03.¹⁹ DLTs included allergic reactions to trial products, new organ toxicities of at least grade 3, or related AEs leading to permanent discontinuation of treatment. AEs resulting in death, hospitalization or life-threatening complications were labeled as SAEs. AESIs included irAEs or cardiac events (cardiac signs or symptoms or clinically significant electrocardiogram [ECG] abnormalities). All SAEs and AESIs were followed to resolution or until the investigator assessed the patient as stable. Safety laboratory tests were performed at screening, the date of first treatment and at each visit with anti-PD-1 dosing. Abnormal values assessed as clinically significant by the investigator were documented as AEs. A screening ECG was performed, with a follow-up ECG performed at week 3 or 4, depending on the cohort. Abnormal ECGs were evaluated by the investigator for clinical significance.

Thirty- and 100-day follow-ups were performed after last treatment to assess any additional safety concerns. Further details pertaining to AE assessment are provided in the Appendix S1.

2.3 |. Clinical assessment

Patients were evaluated in clinic for a targeted physical exam, assessment of Eastern Cooperative Oncology Group (ECOG) performance status, vital signs and routine laboratory tests every 2 weeks ± 4 days through week 15, every 4 weeks ± 7 days through week 51 and every 13 weeks ± 14 days through EOT in C1 and every 3 weeks ± 4 days through week 22, every 6 weeks ± 7 days through week 52 and every 12 weeks ± 14 days through EOT in C2. Baseline tumor assessment was performed within 28 days of the first dose of treatment. Subsequent tumor assessments were performed at 6-week intervals from the first dose of treatment through week 19, at 8-week intervals through week 51 and every 13weeks through EOT in C1, and at 6-week intervals from the first dose of treatment through week 19, at a 9-week interval through week 28 and at 12-week intervals through EOT in C2 using RECIST version 1.1.

2.4 |. Immune correlative studies

PBMCs were collected from phase 1b patients in the safety population who were enrolled at the NCI. Per trial protocol, PBMCs were collected prior to therapy and at weeks 3, 7, 11, 51 and 104 in C1 and prior to therapy at weeks 4, 16, 52 and 100 in C2. Where available, PBMCs were assessed for antigen-specific T-cell responses against the TAAs MUC-1, CEA and brachyury, using a FACS-based assay as described previously.¹⁰

2.5 |. Statistical methods

The phase 1b portion of this trial was designed to enroll at least 12 patients with advanced nonsquamous NSCLC who would receive CV301 in combination with either nivolumab or pembrolizumab. The sample size for phase 1b was not based on statistical power; however, enrollment of 12 patients was considered sufficient to determine the occurrence of DLTs associated with CV301 in combination with anti-PD-1 therapy.

For all trial endpoints, all treated patients defined the safety population and were included for analysis. No treatment comparisons were performed since all phase 1b patients were treated with CV301 and anti-PD-1 therapy. Data summaries were presented by cohort for CV301 in combination with nivolumab or pembrolizumab.

Dosing of phase 1b patients took place sequentially: patients were dosed with at least 2 days between the start of dosing of each patient. If <2 DLTs were observed in the first 6 dosed within 6 weeks of the first dose of CV301 administration, safety was considered established for the combination of CV301 with anti-PD-1 therapy. The safety of phase 1b patients was monitored continuously, with safety monitoring committee meetings held every 6 weeks.

Exploratory endpoints were analyzed exclusively for phase 1b patients in the safety population. OS at 12 months and PFS at 6 and 12 months were based on the proportion of patients meeting the event at each milestone. ORR was determined as the proportion of patients who had a complete response (CR) or partial response (PR) at any time during the trial. Statistical analyses were performed using SAS 9.4 (SAS-Institute, Cary, North Carolina).

3 |. RESULTS

3 |. Patient characteristics

Between October 2017 and September 14, 2018, patients were screened, among whom 12 patients were enrolled and treated and comprised the safety set for analysis. Four patients were treated in C1 and 8 in C2, with the assignment based on prior cancer therapies. Five patients (1 in C1 and 4 in C2) completed the protocol-specified treatment period of at least 1 year of CV301 dosing. The remaining patients withdrew due to adverse events (n = 4) or disease progression (n = 3) (Figure 1).

FIGURE 1.

CONSORT diagram. CONSORT diagram illustrating the number of patients screened, enrolled, treated and included in the safety analysis. Fourteen patients were screened, and twelve patients were enrolled and treated. Patients receiving at least one dose of CV301 with PD-1-directed therapy were included in safety analysis

One patient in each cohort only received the first dose of MVA-BN-CV301, whereas the remaining patients received both planned doses of MVA-BN-CV301 and a median of 11 FPV-CV301 doses (C1 median = 14; C2 median = 11). The median number of nivolumab doses for C1 was 22 (range, 2–36), whereas the median number of doses of pembrolizumab in C2 was 16 (range, 1–23).

Patient characteristics are outlined in Table 1. Of note, the majority of patients had metastatic NSCLC at initial diagnosis (91.7%) and were past smokers (75.0%).

TABLE 1.

Patient characteristics

	Cohort 1 (N = 4)	Cohort 2 (N = 8)	Overall (N = 12)
Age (years)
Median	65.5	63.5	63.5
Min, max	48, 72	57, 74	48, 74
Sex, n (%)
Male	0	3 (37.5)	3 (25.0)
Female	4 (100.0)	5 (62.5)	9 (75.0)
Race group, n (%)
White	2 (50.0)	8 (100.0)	10 (83.3)
Other	2 (50.0)	0	2 (16.7)
Ethnicity, n (%)
Hispanic or Latino	0	1 (12.5)	1 (8.3)
Not Hispanic or Latino	4 (100.0)	7 (87.5)	11 (91.7)
ECOG status, n (%)
0	0	2 (25.0)	2 (16.7)
1	4 (100.0)	6 (75.0)	10 (83.3)
Disease duration at baseline (years)a
Median	1.60	0.45	0.65
Min, max	0.8, 3.2	0.3, 3.1	0.3, 3.2
Stage at initial diagnosis, n (%)
Stage I/II	0	1 (12.5)	1 (8.3)
Locally advanced or stage III	0	0	0
Metastatic or stage IV	4 (100.0)	7 (87.5)	11 (91.7)
Tobacco use, n (%)
Current smoker	0	1 (12.5)	1 (8.3)
Past smoker	3 (75.0)	6 (75.0)	9 (75.0)
Never smoked	1 (25.0)	1 (12.5)	2 (16.7)

Abbreviations: %, n/N total subjects in the cohort; ECOG, Eastern Cooperative Oncology Group; n, number of subjects with a response.

^aDisease duration at baseline in years is calculated as (date of first dose − date of initial diagnosis + 1)/365.25.

Glucocorticoids were used as concomitant medications in seven (58.3%) patients during the trial. Two patients received only topical or inhaled glucocorticoids, one received prednisone prior to scans due to a radiocontrast allergy, three patients continued to use prednisone at a dose of 10 mg orally daily after resolution of irAEs as permitted by protocol and one received dexamethasone as a pre-medication for chemotherapy.

3.2 |. Safety assessment

A summary of treatment-related and ≥grade 3 AEs for the safety population and for each cohort is presented in Table 2. Details of all AEs can be found in Tables S1 and S2.

TABLE 2.

Summary of related and ≥grade 3 TEAEs

	Cohort 1 (N = 4)		Cohort 2 (N = 8)		Overall (N = 12)
System organ class preferred term	Related	≥Grade 3	Related	≥Grade 3	Related	≥Grade 3
Total	4 (100.0)	4 (100.0)	6 (75.0)	3 (37.5)	10 (83.3)	7 (58.3)
General disorders and administrationsite conditions	4 (100.0)	0	6 (75.0)	0	10 (83.3)	0
Injection-site reactionsa	4 (100.0)		5 (62.5)		9 (75.0)
Fatigue	2 (50.0)		1 (12.5)		3 (25.0)
Respiratory, thoracic and mediastinal disorders	2 (50.0)	1 (25.0)	1 (12.5)	0	3 (25.0)	1 (8.3)
Pleural effusion	2 (50.0)				2 (16.7)
Pneumonitisb	1 (25.0)	1 (25.0)	1 (12.5)		2 (16.7)	1 (8.3)
Skin and subcutaneous tissue disorders	1 (25.0)	0	2 (25.0)	0	3 (25.0)	0
Pruritus	1 (25.0)		1 (12.5)		2 (16.7)
Alopecia			1 (12.5)		1 (8.3)
Rash maculo-papular			1 (12.5)		1 (8.3)
Blood and lymphatic system disorders	1 (25.0)	1 (25.0)	1 (12.5)	1 (12.5)	2 (16.7)	2 (16.7)
Anemia			1 (12.5)		1 (8.3)
Disseminated intravascular coagulation	1 (25.0)	1 (25.0)			1 (8.3)	1 (8.3)
Neutropenia				1 (12.5)		1 (8.3)
Gastrointestinal disorders	1 (25.0)	0	1 (12.5)	0	2 (16.7)	0
Diarrhea	1 (25.0)		1 (12.5)		2 (16.7)
Proctitis	1 (25.0)				1 (8.3)
Rectal hemorrhage	1 (25.0)				1 (8.3)
Infections and infestations	0	1 (25.0)	0	1 (12.5)	0	2 (16.7)
Herpes zoster		1 (25.0)				1 (8.3)
Pneumonia				1 (12.5)		1 (8.3)
Injury, poisoning and procedural complications	2 (50.0)	0	0	0	2 (16.7)	0
Infusion related reaction	2 (50.0)				2 (16.7)
Investigations	2 (50.0)	1 (25.0)	0	0	2 (16.7)	1 (8.3)
Body temperature increased	1 (25.0)				1 (8.3)
Weight decreased	1 (25.0)				1 (8.3)
Weight increased		1 (25.0)				1 (8.3)
Musculoskeletal and connective tissue disorders	2 (50.0)	0	0	0	2 (16.7)	0
Arthralgia	1 (25.0)				1 (8.3)
Musculoskeletal chest pain	1 (25.0)				1 (8.3)
Pain in extremity	1 (25.0)				1 (8.3)
Ear and labyrinth disorders	0	1 (25.0)	0	0	0	1 (8.3)
Hypoacusis		1 (25.0)				1 (8.3)
Hepatobiliary disorders	1 (25.0)	1 (25.0)	0	0	1 (8.3)	1 (8.3)
Autoimmune hepatitis	1 (25.0)	1 (25.0)			1 (8.3)	1 (8.3)
Metabolism and nutrition disorders	1 (25.0)	1 (25.0)	0	0	1 (8.3)	1 (8.3)
Decreased appetite	1 (25.0)				1 (8.3)
Hypernatremia		1 (25.0)				1 (8.3)
Nervous system disorders	1 (25.0)	0	0	1 (12.5)	1 (8.3)	1 (8.3)
Cerebrovascular accident				1 (12.5)		1 (8.3)
Headache	1 (25.0)				1 (8.3)
Vascular disorders	1 (25.0)	1 (25.0)	0	0	1 (8.3)	1 (8.3)
Vasculitis	1 (25.0)	1 (25.0)			1 (8.3)	1 (8.3)

Note: Adverse events (AEs) are coded to the MedDRA dictionary version 19.0. Treatment-emergent adverse events (TEAEs) are defined as adverse events occurring after initiation of trial treatment or are present at baseline but worsen in severity after the initiation of trial treatment. Patients reporting a preferred term more than once are counted only once by preferred term and system organ class.

^aInjection-site reactions include injection-site erythema, pain, induration, swelling, mass, nodule and pruritus.

^bOne case of pneumonitis in Cohort 1 was considered a dose-limiting toxicity and serious.

All patients experienced at least one treatment-emergent AE (TEAE), and 10 of 12 patients experienced treatment-related AEs. AEs related to MVA-BN-CV301 or FPV-CV301 were generally mild and included injection-site reactions (erythema, pain, swelling, induration, mass, nodule or pruritus) or fatigue. One patient in C2 discontinued CV301 due to injection-site reactions, including erythema, induration and pain.

All patients in C1 experienced an AE related to anti-PD-1 therapy vs half of C2. The most common AEs related to anti-PD-1 therapy included fatigue (50% in C1 and 12.5% in C2), pleural effusion and pneumonitis (50% in C1 and 12.5% in C2) and skin disorders including pruritus, alopecia and maculo-papular rash (25% each in C1 and C2).

One DLT of pneumonitis was observed in C1 and resulted in discontinuation of treatment. Pneumonitis is a well described side effect of PD-1 inhibitors.¹⁸

Grade 3 or greater AEs and SAEs were observed in 100% and 75% of patients enrolled in C1, and 37.5% and 25% of patients enrolled in C2; however, each event only occurred in one patient such that no event was most frequent. AEs leading to treatment discontinuation were observed in two patients in each cohort, including pneumonitis (also a DLT) and autoimmune hepatitis in C1, and pneumonitis and injection-site reactions in C2.

Overall, the safety profile of CV301 in combination with anti-PD-1 therapy was consistent with the profile for anti-PD-1 therapy alone.^20,21

3.3 |. Adverse events of special interest

In C1, three (75%) patients experienced adverse events of special interest (AESIs), including proctitis, autoimmune hepatitis, pneumonitis, vasculitis and disseminated intravascular coagulation (DIC) (Table 2). All AESIs were considered by the investigator to be related to anti-PD-1 therapy alone, with the exception of pneumonitis which was also considered possibly related to CV301. Proctitis occurred approximately 12 months after initiation of treatment, and autoimmune hepatitis was observed approximately 5 months after treatment initiation. One patient developed pneumonitis, vasculitis and DIC within 2 months of start of treatment. Extensive workup was performed in this case and alternative etiologies, including infectious causes, were ruled out. Despite aggressive management outlined below the patient developed multiorgan failure which was fatal. Pneumonitis, autoimmune hepatitis and proctitis were treated with systemic corticosteroids. DIC and vasculitis required additional immuno-suppression for management, including intravenous immunoglobulins and infliximab.

In C2, one (8%) patient experienced an AESI (pneumonitis) during treatment with pembrolizumab, and 60 days after the last dose of MVA-BN-CV301 and 18 days after the last dose of FPV-CV301. Since a causal relation to CV301, although unlikely, could not be ruled out, pneumonitis was attributed to CV301 and pembrolizumab, required treatment with high-dose corticosteroids and resulted in discontinuation of protocol therapy.

No cardiac events were observed in either cohort (Table 2).

3.4 |. Serious adverse events

SAEs occurred in three (75%) patients in C1 patients, and two (17%) patients in C2. Excluding events characterized as AESIs, encephalitis, dysarthria and memory impairment were observed in one patient in C1. In C2, pneumonia and cerebrovascular accident were reported in addition to previously described AESIs. These non-AESI SAEs that were unrelated to protocol therapy resolved within 1 week and did not lead to discontinuation of protocol therapy.

3.5 |. Clinical activity

Eleven patients were evaluable for response. Overall, two (18%) patients (1 in each cohort) experienced an objective response and nine (82%) patients had stable disease as a best response (Table 3). The responder in C1 had a duration of response of 561 days, and in C2 the response was ongoing at treatment completion with a duration of 92 days at data cutoff (Figure 2). PFS at 12 months for cohorts 1 and 2 was 50% and 25%, respectively (Table 3). OS at 12 months could be assessed in seven patients and was 50% for C1 and 62.5% for C2 (Table 3). Results of genomic profiling were available for two of three evaluable patients in C1, including the patient with an objective response. In both instances the patients’ tumors harbored a KRAS mutation and both patients were progression-free at 12 months.

TABLE 3.

Clinical activity

	Cohort 1 (N = 4)	Cohort 2 (N = 8)
Objective response, n (%)a	1 (25.0)	1 (12.5)
Best overall response, n (%)
Complete response	1 (25.0)	0
Partial response	0	1 (12.5)
Stable disease	2 (50.0)	7 (87.5)
Progressive disease	0	0
Inevaluable	1 (25.0)	0
Duration of response (days)
n	1	1
Mean (SD)	561.0 (NE)	92.0 (NE)
Median	561.0	92.0
Min, max	561, 561	92, 92
Overall survival at 12 months, n (%)	2 (50.0)	5 (62.5)
Deaths prior to 12 months, n (%)	1 (25.0)	0
Censored due to early withdrawal,b n (%)	1 (25.0)	3 (37.5)
Progression free survival, n (%)
6 months	2 (50.0)	6 (75.0)
12 months	2 (50.0)	2 (25.0)

Abbreviations: %, N total subjects in the cohort; n, number of subjects with a response; NE, not estimable.

^aObjective response is defined as a best overall response of complete response or partial response based on investigator assessment using RECIST v1.1.

^bCensored patients had <12 months of follow-up.

FIGURE 2.

Change in tumor burden over time by best objective response. Cohort 1: Four patients entered Cohort 1; however, only three had evaluable post-baseline tumor evaluations. One patient developed pneumonitis with associated radiological changes due to which post-baseline tumor evaluation could not be performed. Therefore, only three patients are represented in the figure. Cohort 2: One patient in Cohort 2 developed a moderate pleural effusion on treatment (represented as a “new lesion”) that was determined to represent unequivocal disease progression by the investigator

3.6 |. Immunological studies

PBMCs were available to measure antigen-specific T-cell responses prior to and during therapy from three patients in C1, and one patient in C2 who were treated at the NCI. The FACS-based assay for T cells expressing the type I cytokines IFNγ, IL2, TNFα and/or the degranulation marker CD107a following stimulation with overlapping peptide pools has been previously described.¹⁰ All patients evaluated developed T-cell responses after therapy to at least one of the antigens tested; four (100%) patients developed MUC-1 responses, three (75%) patients developed CEA responses and two (66.7%) patients developed T cells against the cascade antigen brachyury (Table S3). Multifunctional T cells, defined as CD4⁺ or CD8⁺ T cells that express two or more of the markers examined, were also measured. Using a cutoff of a >3-fold increase after therapy, all patients developed multifunctional T cells to at least one of the antigens tested, with three (75%) patients developing multifunctional T cells to MUC-1 and CEA and four (100%) patients developing multifunctional T cells to brachyury (Table 4). One patient developed a >10-fold increase in multifunctional MUC-1- and brachyury-specific T cells, and a >100-fold increase in multifunctional CEAspecific T cells after therapy.

TABLE 4.

Multifunctional TAA T cell responses pre- and posttreatment

		MUC1		CEA		Brachyury
Patient (cohort)	Days posttreatment	2 or more CD4	2 or more CD8	2 or more CD4	2 or more CD8	2 or more CD4	2 or more CD8
1 (C1)	D1	423	0	242	242	0	1116
	D14	0	193 (+)	8	8	59	73
	D43	179	179 (+)	490	740 (+)	971 (+)	773
	D71	0	36	389	0	112 (+)	251
2 (C1)	D1	338	589	66	233	144	228
	D14	532	208	256 (+)	583	83	164
	D43	349	0	236 (+)	375	3123 (++)	347
	D155	121	0	0	753 (+)	1710 (++)	1341 (+)
3 (C1)	D1	342	59	28	16	51	0
	D14	256	1904 (++)	99	207 (++)	30	90
	D43	399	5270 (++)	52	5114 (+++)	314 (+)	2946 (++)
5 (C2)	D1	762	21 478	598	13 424
	D21	582	4073	0	24
	D106	5413 (+)	3547	0	0

Note: Multifunctional immune responses against the TAAs MUC-1, CEA and brachyury were defined as CD4⁺ and CD8⁺ T cells expressing two or more among IFNγ, TNFα, IL2 or CD107a, following stimulation of PBMCs with overlapping peptide pools. PBMCs were collected from patients before (D1) and at the indicated time points during therapy. The absolute number of multifunctional cells per 1 × 10⁶ PBMCs plated at the start of the assay was calculated and background (obtained with the negative control peptide pool, HLA) was subtracted. Patients were scored as having developed a low (+, >3× pre, or if no pre >100/1 × 10⁶ cells post), mid (++, >10× pre, or if no pre >1000/1 × 10⁶ cells post) or high (+++, >100× pre, or if no pre >5000/1 × 10⁶ cells post) multifunctional T-cell response after therapy. Values in bold indiate multifunctional T cell responses after therapy. Gray box indicates that insufficient PBMCs were available for analysis of brachyury-specific T cells in patient 5 (C2).

Abbreviations: C1, Cohort 1; C2, Cohort 2; CEA, carcinoembryonic antigen; D, day; IFNγ, interferon gamma; MUC-1, mucin-1; TAA, tumor-associated antigen; TNFα, tumor necrosis factor alpha.

4 |. DISCUSSION

Our trial demonstrates the safety and clinical activity of CV301 in combination with PD-1-directed immunotherapy. Treatment was generally well tolerated, and most AEs were manageable. The irAE profile of CV301 in combination with anti-PD-1 therapy was similar to anti-PD-1 monotherapy. An exception to this pattern was observed in a patient with serological evidence of a pre-existing connective tissue disorder, but no clinical history of active autoimmune disease, who experienced severe injection-site reactions and subsequently developed pneumonitis, vasculitis and coagulopathy, possibly due to an underlying predisposition to immune-mediated toxicity. Differences in the proportion of patients experiencing treatment-related AEs, including AESIs in cohorts 1 and 2, are likely a result of the small number of participants in each cohort and do not appear to be related to the type of anti-PD-1 antibody used in combination with CV301 or to differences in patient characteristics such as age, sex or ethnicity. As mentioned above, it is conceivable that one of four patients in C1 had a predisposition to irAEs. Another patient in C1 experienced an SAE that was unrelated to treatment.

The clinical activity of the combination of CV301 and anti-PD-1 therapy in patients with NSCLC is encouraging in this small trial. However, larger trials are required to confirm this observation. C1, consisting of patients receiving CV301 in combination with nivolumab as secondline treatment for nonsquamous NSCLC, had 6- and 12-month PFS of 50% each. Of note, two of three evaluable patients with durable benefit in C1 had tumors harboring a KRAS mutation. Although it is not possible to interpret the significance of a KRAS mutation on treatment outcome in this limited subset of patients, it should be noted that a similar signal of activity was observed in patients with KRAS-mutated tumors treated with CV301 alone in the phase 1a, dose-escalation portion of this trial.¹⁰ C2, which consisted of nonsquamous NSCLC patients receiving frontline treatment with pembrolizumab and CV301, demonstrated a 6- and 12-month PFS of 75% and 25%, respectively. The limited number of patients enrolled in this trial and absence of a control arm make it difficult to assess clinical benefit relative to anti-PD-1 therapy alone. Nevertheless, in a phase 3 trial comparing nivolumab monotherapy with docetaxel in a similar patient population, 12-month PFS in the nivolumab arm was 19%.²² Similarly, in a phase 3 trial comparing pembrolizumab with chemotherapy in patients with untreated advanced NSCLC with tumor cell PD-L1 expression of 50% or higher and no actionable mutations, the 6-month PFS in the pembrolizumab arm was 62%.¹⁷ Our results compare favorably with these observations.

Immunologic analyses, although performed in a small number of patients, demonstrate that CV301 administered in combination with PD-1 checkpoint blockade is immunogenic, inducing both MUC-1-and CEA-specific T cells, including multifunctional T cells, in the majority of patients evaluated. In addition, multifunctional T-cell responses against brachyury, an antigen not encoded by the vaccine, were also observed after treatment, suggesting that the regimen of CV301 with PD-1 blockade induces immunologically relevant tumor cell destruction.

In conclusion, the combination of CV301 and anti-PD-1 therapy is safe and appears clinically active in patients with advanced nonsquamous NSCLC. Treatment is associated with generation of multifunctional TAA T-cell responses. These results support further evaluation of cancer vaccines in combination with ICIs to augment the antitumor immune response generated by either intervention alone in patients with advanced NSCLC.

DATA AVAILABILITY STATEMENT

The datasets used and/or analyzed during the current trial are available from the corresponding author on reasonable request.

What’s new?

CV301 is a poxviral-based cancer vaccine that generates antigen-specific T-cell responses to MUC-1 and CEA, which are overexpressed in many non-small cell lung cancers. We tested the hypothesis that CV301 in combination with programmed death-1-directed therapy is safe and immunologically active. Our trial demonstrates the tolerability and clinical activity of CV301 with nivolumab or pembrolizumab in patients with advanced non-small cell lung cancer and its ability to generate multifunctional immune responses to MUC-1 and CEA.

Abbreviations:

AE

adverse event

AESI

adverse event of special interest

ALK

anaplastic lymphoma kinase

C1

cohort 1

C2

cohort 2

CEA

carcinoembryonic antigen

CR

complete response

CRADA

Cooperative Research and Development Agreement

CTCAE

Common Terminology Criteria for Adverse Events

D

day

DIC

disseminated intravascular coagulation

DLT

dose limiting toxicity

DOR

duration of response

ECG

electrocardiogram

ECOG

Eastern Cooperative Oncology Group

EGF

epidermal growth factor

EGFR

epidermal growth factor receptor

EOT

end of treatment

FPV

Fowlpox-CV301

HLA

human leukocyte antigen

ICAM-1

intercellular adhesion molecule-1

ICI

immune checkpoint inhibitor

IFN-γ

interferon-gamma

Inf.U

infectious units

irAE

immune-related adverse event

IRB

Institutional Review Board

IV

intravenous

LFA-3

leukocyte function-associated antigen-3

MUC-1

mucin-1

MVA

modified vaccine Ankara-BN-CV301

NCI

National Cancer Institute

NE

not estimable

NSCLC

non-small cell lung cancer

ORR

objective response rate

OS

overall survival

PBMC

peripheral blood mononuclear cell

PD-1

programmed death-1

PD-L1

programmed death-ligand 1

PFS

progression-free survival

PR

partial response

RECIST

Response Evaluation Criteria in Solid Tumors

SAE

serious adverse event

TAA

tumor-associated antigen

TEAE

treatmentemergent adverse event

TNF-α

tumor necrosis factor-alpha