Abstract
Purpose: Intratumorally injected Clostridium novyi-NT (nontoxic; lacking the alpha toxin), an attenuated strain of C. novyi, replicates within hypoxic tumor regions resulting in tumor-confined cell lysis and inflammatory response in animals, which warrants clinical investigation.
Patients and Methods: This first-in-human study (NCT01924689) enrolled patients with injectable, treatment-refractory solid tumors to receive a single intratumoral injection of C. novyi-NT across 6 dose cohorts (1 × 104 to 3 × 106 spores, 3+3 dose-escalation design) to determine dose-limiting toxicities (DLT), and the maximum tolerated dose.
Results: Among 24 patients, a single intratumoral injection of C. novyi-NT led to bacterial spores germination and the resultant lysis of injected tumor masses in 10 patients (42%) across all doses. The cohort 5 dose (1 × 106 spores) was defined as the maximum tolerated dose; DLTs were grade 4 sepsis (n = 2) and grade 4 gas gangrene (n = 1), all occurring in three patients with injected tumors >8 cm. Other treatment-related grade ≥3 toxicities included pathologic fracture (n = 1), limb abscess (n = 1), soft-tissue infection (n = 1), respiratory insufficiency (n = 1), and rash (n = 1), which occurred across four patients. Of 22 evaluable patients, nine (41%) had a decrease in size of the injected tumor and 19 (86%) had stable disease as the best overall response in injected and noninjected lesions combined. C. novyi-NT injection elicited a transient systemic cytokine response and enhanced systemic tumor-specific T-cell responses.
Conclusions: Single intratumoral injection of C. novyi-NT is feasible. Toxicities can be significant but manageable. Signals of antitumor activity and the host immune response support additional studies of C. novyi-NT in humans.
Translational Relevance
Our findings suggest that single intratumoral injection of Clostridum novyi-NT is feasible, and the effects of bacterial spores germination are manageable. Early signals of antitumor activity and the host immune response were observed and support additional studies of Clostridium novyi-NT in humans.
Introduction
More than 100 years ago, Coley's discovery that postsurgical infection can help control cancer led to clinical testing of the therapeutic use of Streptococcus pyogenes (Coley toxin) in patients with sarcomas (1). However, limited supportive care resources to combat the anticipated adverse events (AEs) resulted in this approach being marginalized. The contemporary development of a plethora of new immunotherapy approaches has led to renewed interest in the investigation of viruses and bacteria, as anticancer agents owing to their ability to elicit immune anticancer responses (2).
Clostridium novyi-NT (nontoxic) is an attenuated strain of C. novyi, a spore-forming, gram-positive, obligate anaerobe that lacks the lethal alpha toxin (3). When administered intravenously or intratumorally, C. novyi-NT spores colonize and replicate exclusively within the hypoxic regions of tumors, eliciting robust, precise, and tumor-confined cell lysis and immune inflammatory response resulting in immunogenic cell death in animals, including companion dogs with naturally occurring tumors (3–6). The mechanisms by which C. novyi-NT induces tumor destruction are incompletely understood but may include direct tumor cell lysis as well as the activation of the host immune system (4). The development of C. novyi-NT for use in humans has been largely driven by the results of companion dog studies, which revealed better tolerability and higher objective response rates (ORR) with intratumoral administration (ORR, 40%) than with intravenous administration (ORR, 9%; refs. 5, 6). Those observations supporting anticancer activity with intratumoral administration and the management of common AEs, such as fever, tumor inflammation, and tumor abscessation (which were expected owing to the nature and mechanisms of C. novyi-NT treatment) provided the basis for further investigation in human trials including toxicity risk-mitigation strategies.
We hypothesized that intratumoral administration of C. novyi-NT can have anticancer activity by inducing the tumor-confined cell lysis and host immune and inflammatory response resulting immunogenic cell death. In that context, we undertook an open-label, first-in-human phase I study (Clinicaltrials.gov identifier, NCT01924689) of a single, intratumoral injection of C. novyi-NT to evaluate its safety profile, maximum tolerated dose (MTD), dose-limiting toxicities (DLTs), preliminary antitumor activity, and host immune and inflammatory response in patients with treatment-refractory solid tumors.
Patients and Methods
Patients
Eligible patients were ≥18 years old and had a diagnosis of treatment-refractory solid tumor with a target tumor that was measurable, palpable, or clearly identifiable under radiographic guidance and amenable to percutaneous injection of C. novyi-NT spores (Supplementary File S1). The target tumor for injection was required to have a longest diameter of ≥1 cm and ≤12 cm and not be located in the thoracic, abdominal, or pelvic cavity or in the brain. After the first patient was treated, the study was amended to exclude patients with clinical, functional, or radiographic evidence of bone involvement at the site of the target lesion for injection. Additional inclusion criteria were an Eastern Cooperative Oncology Group (ECOG) performance status score of ≤2 and adequate bone marrow, renal, hepatic, and cardiac function. Patients with conditions that might limit the ability to eradicate a C. novyi-NT infection, including asplenia, concomitant immunosuppressive medication, or multiple antibiotic allergies, were excluded.
Study design and treatment
This open-label study was conducted at six centers in the United States. Eligible patients were enrolled sequentially into one of six dosing cohorts using a 3+3 dose-escalation design (Table 1). The primary objective of this phase I study was to determine the safety profile, DLTs, and MTD of C. novyi-NT. DLTs were defined as any adverse event (AE) of grade 3 toxicity lasting 3 days or longer, or any AE of grade 4 toxicity of any duration that was assessed to be at least possibly related to C.novyi-NT and occurred with 4 weeks from C. novyi-NT administration (DLT criteria are detailed in the protocol; Supplementary File S1). Key secondary endpoints included the antitumor activity of C. novyi-NT in the injected tumor, overall response to C. novyi-NT as assessed by RECIST 1.1, and the immune and inflammatory responses induced by C. novyi-NT. The first patient in a cohort was observed for a minimum of 2 weeks before treating the second patient, while there was no required minimum observation interval before subsequent patients in the same cohort could be dosed. The starting dose was 1 × 104 spores administered as a single intratumoral injection, and doses were increased in successive cohorts until MTD was reached.
Dose levels and Dose-limiting toxicities.
The study protocol was approved by the Institutional Review Board at each participating site, and all patients provided written informed consent before their inclusion in the study. The study was conducted in accordance with local legal requirements, with the U.S. Code of Federal Regulations (21 CFR 312.50 and 21 CFR 56), and with the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use guidelines on good clinic practice and the moral, ethical, and scientific principles governing clinical research as set out in the Declaration of Helsinki (version 2008).
C. novyi-NT spores were manufactured and formulated by Omnia Biologics. The investigational product was prepared on the day of its administration to the patient. The dilution of the concentrated spore suspension was performed in a designated biological safety cabinet using sterile saline (0.9%) infusion bags of appropriate sizes to achieve the required dose based on the dose levels of the respective cohorts. The injection volume (3 mL) was withdrawn from the saline bag and injected percutaneously with or without radiographic guidance (CT or ultrasound) with either a 22- to 24-gauge needle or, in five patients, a Quadra-Fuse multipronged 18-gauge needle (Rex Medical).
Patients remained hospitalized after C. novyi-NT spores injection and were observed for 8 days for signs and symptoms of C. novyi-NT spores germination and AE. C. novyi-NT spores germination was defined as an intratumoral infection resulting in tumor lysis and a constellation of clinical signs and symptoms, including fever, injection site pain, erythema, swelling, tenderness, and in some cases, ulceration, spontaneous drainage, tissue sloughing, bleeding, and malodor. On day 8, patients were required to start oral doxycycline (100 mg twice a day) to prevent the additional germination of C. novyi-NT spores. Clinical management, including the decision to use antibiotics before day 8, was guided by an algorithm that incorporated the SOFA tool, with scores above 8 predicting mortality rates averaging 55% (range, 26%−95%) and scores below this threshold predicting mortality rates averaging 13% (range 0%−21%; refs. 7–9). Following discharge from the hospital, patients were evaluated in the outpatient clinic twice weekly for 3 weeks. Subsequent clinic visits, which included safety and efficacy assessments, occurred 2, 4, 8, and 12 months after discharge. Clinical responses were evaluated using CT, MRI, or clinical examination using RECIST version 1.1. (10). The density of the injected tumors before and after C. novyi-NT injection was measured using the method described in the Choi criteria (11).
Assessment of inflammatory and immunologic characteristics
Cytokine responses
Venous blood was drawn from the patients at baseline and after treatment (at hour 1, hour 12, days 1–7, day 14, day 21, month 2, month 4, and month 8) and analyzed for serum cytokine levels using Ciraplex Assay (CiraTM Custom Kits, Aushon BioSystems). Differences between pretreatment and posttreatment cytokine concentrations were calculated. The ratio of the difference was used as a measure of the concentration change for each protein at each time point in each patient. Peak response dates for the cytokines were computed. Relative differences of all cytokines were compared for patients with and without clinical germination. Days 3 and 5 were identified as having the most elevated cytokines within the first week after C. novyi-NT injection. Results are presented as bar plots showing group means, SEs, and statistical significance between groups.
Systemic tumor antigen−specific T-cell responses
Venous blood samples were collected from patients at baseline (day 0) and after the C. novyi-NT injection (days 7, 14, and 21; months 1, 2, and 8 if feasible). Peripheral blood mononuclear cells (PBMC) were isolated by density gradient centrifugation as described previously (12). Tumor cell lines that matched the cancer of each patient were used to prepare cell extracts, which then served as sources of allogenic tumor antigens. Pretreatment and posttreatment PBMCs from each patient were plated at 0.5–1 × 106 cells per well and incubated at 37°C in a 5% CO2 atmosphere with CEF Control Peptide Pool “Plus” (CTL; positive control) or PBMC lysate at a final concentration of 10 μg/mL (negative control). The frequencies of tumor antigen−specific T cells within PBMCs were determined by enzyme-linked immunosorbent spot (ELISPOT) assays measuring the release of T-cell cytokines such as IFNγ, granzyme B, and TNFα (13). A positive response for T-cell activation was defined as a T-cell frequency that was detectable (i.e., >1:100,000) and significantly (P < 0.05, two-tailed t test) greater than the mean T-cell frequency of the control no-antigen wells. Differences in the pretreatment and posttreatment frequencies of tumor antigen−specific T cells were analyzed with the paired Student t test.
Local tumor-specific T-cell responses
Core-needle biopsy specimens of the injected tumor and one noninjected lesion were collected at baseline, month 1 (optional), and month 2. Three 4-μmol/L sections from each formalin-fixed, paraffin-embedded biopsy tissue block were subjected to immunofluorescence staining using antibodies against human CD8 (Abcam, ab4055), CD14 (Abcam, ab45870), CD15 (Abcam, ab135377), CD68 (Thermo Fisher Scientific, MA5-12407), perforin (Abcam, ab75573), granzyme B (eBioscience, 14-8889), caspase 3 (Cell Signaling Technology, 9661S), FOXP3 (Abcam, ab20034), RORgt (eBioscience, 14-6988), and CD3 (Thermo Scientific Pierce, MA1-90582). Two investigators independently analyzed the stains using an Axio Observer Z1 inverted fluorescence microscope (Zeiss). Staining was scored by counting the stained cells in 5–20 fields of vison (magnification ×200) in three slides per tumor biopsy specimen and using a modified Allred scoring method (1/100 of counted cells = 0–1+; 1/10 of counted cells = 2+; 1/3 of counted cells = 3+; 2/3 of counted cells = 4+; almost 100% = 5+). Respective counts were averaged in each case.
Cell lines for extracts used in immunospot
Breast cancer: MCF7 (Breast), KS22.24 (Breast). Colon cancer: HCT 116 (ATCC CCL-247, Colon) SW480 (ATCC CCL-228, Colon). Osteosarcoma: (Dr. Maran Avudaiappan -Avudai, Mayo Clinic) 143B (Osteosarcoma-metastatic): KH05 (Osteosarcoma-nonmetastatic); MG63 (Osteosarcoma-nonmetastatic), LM7 (Osteosarcoma-metastatic). Ovarian cancer: (Dr. Scott Kaufmann, Mayo Clinic), CaOV3: (Ovarian adenocarcinoma), CaOV4: (Ovarian adenoca-metast.), Tyk-nu: (Ovarian, undifferentiated carcinoma). Melanoma: (Dr. Svetomir Markovic, Mayo Clinic), A-375 (ATCC CRL-1619 Melanoma), C32 (ATCC CRL-1585 Melanoma) SK-Mel-28 (ATCC HTB-72 Melanoma). Chordoma: U-CH-1→(ATCC CRL-3217), MUG-Chor1→ (ATCC CRL-3219), U-CH-2→(ATCC CRL-3218). Endometrial adenocarcinoma: (Dr. Gaurang Daftary, Mayo Clinic), Ishikawa (Endometrial adenocarcinoma- Sigma 99040201), HEC-1-A: (Endometrial adenocarcinoma- ATCC HTB-112), HEC1-B: (endometrial adenocarcinoma- ATCC HTB-113), KLE: (uterus/endometrial adenocarcinoma- ATCCCRL1622). Spindle cell sarcoma: Hs 132.T (spindle cell sarcoma- ATCC CRL-7085), Hs 63.T (dermatofibrosarcoma protuberans- ATCC CRL-7043), Hs 295.Sk (dermatofibrosarcoma- ATCC CRL-7232). Pelvic sarcoma: (EMS): H-EMC-SS (extraskeletal myxoid chondrosarcoma- Sigma cat. No. 94042258), Hs 819.T (bone chondrosarcoma- ATCC CRL-7891), SW 1353 (bone chondrosarcoma- ATCC HTB-94), C-20/A4 (human chondrocyte, articular cartilage- EMD- Millipore cat. No. SCC041). Leiomyosarcoma: SK-LMS-1 (vulva, fibroblast- ATCC HTB-88), SK-UT-1B (uterus; endometrium, grade III, mesodermal tumor (mixed), consistent with leiomyosarcoma- ATCC HTB-115), Hs 5.T (connective tissue, leiomyosarcoma- ATCC CRL-7822). Oropharyngeal tumor lysate (Tumor biopsy from tissue bank of Dr. Jan Kasperbauer, Mayo Clinic).
Statistical analysis
Progression-free survival duration was defined as the time from the initiation of systemic therapy to the date of disease progression or death from any cause. The Mann–Whitney U test was applied to assess the association between the size of injected lesion and germination. The Fisher exact test was used to assess association between the frequency of ≥ grade 3 AEs and germination. All tests were two-sided, and P values <0.05 were considered statistically significant. Statistical analyses were performed using the SPSS 24 (SPSS) software program.
Results
Demographics and disease characteristics
Between November 2013, and April 2017, 24 patients (12 men and 12 women) with advanced refractory solid tumors were enrolled to six dose levels of single intratumoral injection of C. novyi-NT spores (Table 1). Eligibility criteria are summarized in Patients and Methods and detailed in the protocol (Supplementary File S1). The median patient age was 55 years (range, 29–75 years); most patients (n = 13, 54%) had sarcomas. Patients' baseline demographics and disease characteristics are shown in Table 2. The sizes and sites of the injected tumors are detailed in Table 3.
Patient and clinical characteristics of 24 patients at baseline.
Outcomes of the 24 patients.
Safety evaluation
All 24 patients enrolled across six dose levels were evaluable for safety and DLTs [DLT criteria summarized in Patients and Methods and detailed in the protocol (Supplementary File S1)]. A total of 3 DLTs, which included grade 4 sepsis at dose levels 4 and 6 and grade 4 gas gangrene at dose level 6, were noted in three patients (Table 1). All patients recovered from their DLTs as described in detail below. On the basis of the finding of two DLTs in dose level 6 (3 × 106 C. novyi-NT spores), the MTD was determined to be 1 × 106 C. novyi-NT spores (dose level 5).
The first DLT was grade 4 sepsis, reported in a 37-year-old man with metastatic osteosarcoma who received an injection of 3 × 105 C. novyi-NT spores (dose level 4) into a 10.9-cm lesion at the stump of his right leg amputation (Table 3). The patient developed a fever of 38.9°C within 8 hours after injection. Three days after injection, the patient progressively developed sepsis with persistent fever (39.2°C), hypotension (65/37 mmHg), tachycardia (162 beats/minute), thrombocytopenia (84 × 103 platelets/μL), and a maximum Sequential Organ Failure Assessment (SOFA, detailed in Patients and Methods and Supplementary File S1) score of 8. CT demonstrated gas pockets (Fig. 1A). The patient received intravenous antibiotics, hydration, and vasopressors and recovered from sepsis without sequelae; however, the patient died 56 days after injection owing to disease progression at both the injected and noninjected sites.
Serial CT, MRI, and photography of injected tumors in patients with Clostridium novyi-NT germination. A, Serial CT of a Clostridium novyi-NT−injected lesion (red arrows) in the lower extremity amputation area of a patient with osteosarcoma demonstrates pockets of gas. B, Serial CT of an injected lesion in the right hip of a patient with leiomyosarcoma demonstrates replacement of the tumor with gas and necrotic material. C, Serial T2-weighted MRI of an injected lesion in the arm of a patient with myxofibrosarcoma of the right arm demonstrates diffuse gas accumulation in subcutaneous, intraosseous, and intramuscular space and interval decrease in size of complex mass. D, Serial CT of an injected lesion (top) in the abdominal wall of a patient with liposarcoma demonstrates near-complete replacement of the tumor mass with gas and necrotic material; the noninjected lesion (bottom) in the abdominal wall shows no progression. E, Serial MRI of an injected lesion in the area of the upper humerus in a patient with leiomyosarcoma demonstrates significant reduction of contrast enhancement; persistent enhancement in the periphery of the tumor, which likely had less hypoxia, is indicated with yellow arrows. F, Serial CT of an injected lesion in the abdominal wall of a patient with uterine carcinosarcoma demonstrates small pockets of gas and necrotic material. G, Serial MRI of an injected lesion in the neck of a patient with angiosarcoma demonstrates necrotic material replacing the center of the tumor. H, Serial photography of an injected lesion in the chest wall of a patient with breast carcinoma demonstrates swelling. I, Serial CT of an injected lesion in the axilla of a patient with adenocarcinoma of unknown primary demonstrates gas and necrotic material replacing the tumor. Yellow arrow indicates a drainage catheter which was placed to facilitate drainage of the necrotic mass. J, Serial MRI of an injected lesion in the arm of a patient with undifferentiated pleomorphic sarcoma demonstrates significant reduction in contrast enhancement.
The second DLT was grade 4 sepsis in a 65-year-old man with leiomyosarcoma who received an injection of 3 × 106 C. novyi-NT spores (dose level 6) into an 8.5-cm right anterior thigh mass (Table 3). The patient developed a fever of 39.4°C 24 hours after injection and progressed to sepsis by day 6 with a maximum SOFA score of 4. On day 6 after injection, CT revealed gas accumulation without tracking along fascial planes and adjacent inflammatory changes (Fig. 1B). The patient was treated with intravenous hydration, antibiotics, and supportive care, which resulted in a full recovery by day 12. At month 2, the assessment showed stable disease (SD) in the injected lesion but overall disease progression; the patient died 4.5 months after C. novyi-NT injection owing to disease progression.
The third DLT was grade 4 gas gangrene in a 66-year-old man with myxofibrosarcoma of the right upper extremity. The patient was injected with 3 × 106 C. novyi-NT spores (dose level 6) into an 8.1-cm tumor in the right arm, which was associated with grade 3 lymphedema. The patient developed increased pain, paresthesia, and discoloration, and MRI revealed diffuse gas accumulation in subcutaneous, intraosseous, and intramuscular spaces and necrosis without sepsis (Fig. 1C). Despite intravenous antibiotics and supportive care, the patient's condition worsened; 11 days after administration, owing to ischemia and gangrene, the patient underwent right upper extremity amputation, the only standard therapeutic option. Blood cultures were positive for C. novyi-NT on days 3 and 5. Postamputation tissue cultures demonstrated coagulase negative Staphylococcus, Corynebacterium spp and C. novyi. The patient was discharged 25 days after C. novyi-NT injection and remained progression-free for 4 months after C. novyi-NT injection.
C. novyi-NT germination was defined as an intratumoral infection resulting in tumor lysis and a constellation of clinical signs and symptoms, including fever, injection site pain, erythema, swelling, tenderness, and in some cases, ulceration, spontaneous drainage, tissue sloughing, bleeding, and malodor (Table 4). Laboratory findings, such as elevated C-reactive protein (CRP) levels and white blood cell counts, and radiographic findings evident of abscess formation, including gas pockets and air fluid levels, were observed. In some patients, CT-guided percutaneous drain placement into the tumor abscess was undertaken to facilitate drainage and bacteriologic examination. C. novyi-NT germination was observed in 10 patients (42%), across every dose cohort and across the spectrum of tumor histologies (Table 3). Germination was observed across all dose levels. In addition, germination occurred across a range of tumor sizes (2.6–11.9 cm); patients with germination had a significantly larger median injected tumor size [7.9 cm; 95% confidence interval (CI), 5.4–9.7] than those without germination did (3.1 cm; 95% CI, 2.9–5.0; P = 0.006). In addition to DLTs, C. novyi-NT−related grade 3 or 4 AEs included pathologic fracture (n = 1), soft-tissue infection (n = 1), respiratory insufficiency (n = 1), limb abscess (n = 1), and rash (n = 1), which all occurred in four patients with germination. Collectively, eight episodes of grade ≥3 AEs (DLTs and others) were observed in five of the 10 patients with germination (50%) and in none of the 14 patients without germination (P = 0.006). Of interest, all five patients with grade ≥3 AEs had larger tumors (4.3, 7.6, 8.1, 8.5, and 10.9 cm, respectively), and all three patients with DLTs had tumors ≥8 cm (10.9, 8.5, and 8.1 cm, respectively). Finally, regular blood cultures detected C. novyi-NT in blood samples from four patients (three with germination and one without germination) collected during the first week of therapy.
Treatment-related AEs, including DLTs.
Efficacy
The median duration of study participation, starting from the day of injection, was 1.9 months (range, 0.4–8.2 months). At the time of the analysis, all 24 patients had discontinued the study: 12 patients without disease progression (50%) discontinued because they started a new systemic therapy; one (4%) because of starting radiotherapy, seven (29%) because of disease progression; two (8%) because they died from disease progression; and two (8%) because they withdrew consent.
Of the 24 patients, 10 (42%) had radiological and clinical signs of tumor destruction from C. novyi-NT germination in the injected lesions, and some of these local responses were rapid and dramatic (Fig. 1A–J). Of the 22 evaluable patients with adequate imaging assessments, nine (41%) had posttreatment CT and/or MRI that showed a decrease in the size of the injected lesion (−2% to −24%) compared with baseline (Table 3; Fig. 2).
Best percentage change in the sum of the largest diameters of target injected and noninjected lesions (gray), largest diameter of the injected lesion (blue), and sum of the largest diameters of target noninjected lesions (red) for 22 evaluable patients.
Among the 22 evaluable patients, the overall RECIST 1.1 assessments of the injected and noninjected lesions showed that 19 (86%) had SD and three (13%) had progressive disease as the best response. Of the 19 patients with SD, 10 had SD on imaging at 4 weeks and nine had SD at 4, 8, or more weeks (Table 3).
We also analyzed changes in the density of the injected lesions on contrast-enhanced CT using the methodology defined by the Choi criteria (Supplementary Fig. S1).
Host immune and inflammatory response to C. novyi-NT
Cytokine responses
Changes in circulating cytokine levels in response to therapy were measured in all 24 patients. Cytokine changes started as early as 12 hours after injection of C. novyi-NT and typically peaked within 6 days. Compared with patients without signs of germination, those with signs of germination had significantly higher cytokine levels (P < 0.05; Fig. 3A and B). The elevated cytokines included IL6 and CRP; IL18; IL10; macrophage colony-stimulating factor; the lymphocyte survival factor IL15; VEGF and its receptor VEGFR1; hepatocyte growth factor; and matrix metalloproteinase-9 and tissue inhibitor of metalloproteinase-1. Overall, these observations are consistent with rapid acute antimicrobial innate immune responses followed by damage-induced tissue repair and regeneration.
Host immune and inflammatory response to C. novyi-NT. A, Changes in the relative concentrations of the top five circulating analytes on day 3 after injection compared with baseline. B, Changes in the relative concentrations of the top nine circulating cytokines on day 5 after injection compared with baseline. C, T-cell cytokine responses in PBMCs in patients without clinical germination. D, T-cell cytokine responses in PBMCs in patients with clinical germination. E, Scoring of immunofluorescence staining. Yellow indicates increased infiltration scores 4–8 weeks after treatment. All three patients had clinical germination and positive ELISPOT results. Scoring was performed using the modified Allred scoring method (1/100 of counted cells = 0–1+; 1/10 of counted cells = 2+; 1/3 of counted cells = 3+; 2/3 of counted cells = 4+; almost 100% = 5+). MDSCs were defined as CD68+/CD14+ for monocytic MDSCs and CD68+/CD15+ for granulocytic MDSCs. Monocytes were defined as CD14+/CD68−/CD15−. F, Immunofluorescence staining of T cells expressing CD8 (green), perforin (red), or granzyme B (GrzB; white) at baseline and at week 4/week 8 in injected lesions from three selected patients with germination (magnification ×400).
Systemic tumor antigen−specific T-cell responses
Tumor antigen stimulated release of T-cell IFNγ, TNFα, and granzyme B, were simultaneously measured following the pulse treatment of PBMCs with specific tumor antigens (14). Of the 12 patients for whom serial PBMC samples were available, seven demonstrated tumor-specific T-cell responses starting as early as 1 week after injection; in five of these patients, the responses lasted for 8 weeks after C. novyi-NT injection (Fig. 3C and D; Supplementary Fig. S2). Concomitant tumor-specific IFNγ and TNFα responses were detected in three of three tested patients with clinical germination but in only one of four patients without clinical germination (Fig. 3C and D). IFNγ responses with TNFα or granzyme B responses were observed in two of four patients who had no clinically detectable germination.
Local tumor-specific T-cell responses
Pretreatment and posttreatment (week 4–8) tumor biopsy specimens were obtained from eight patients, of whom five had germination and three did not. All tumors except one had tumor-infiltrating lymphoid and myeloid cells prior to C. novyi-NT injection. Among the five patients with germination, three had increased infiltration of T cells [comparable for effector cells (T-cell and perforin] and suppressor cells [Treg (CD8−/CD3+/FoxP3+), myeloid-derived suppressor cell (MDSC; CD68+/CD14+ for monocytic MDSC; CD68+/CD15+ for granulocytic MDSC)], whereas none of the patients without clinical germination had such infiltration (Fig. 3E and F). In three patients (two with and one without germination) for whom paired biopsy specimens of injected and noninjected lesions were available, we did not detect any increase in tumor infiltration in the noninjected lesions; however, these three patients also did not have any increase in tumor infiltration in the injected lesions (Supplementary Fig. S3).
Discussion
In this open-label, multicenter, phase I study, a single intratumoral injection of C. novyi-NT was proven to be feasible with an MTD of 1 × 106 spores. DLTs and grade ≥3 AEs related to C. novyi-NT germination were deemed to be manageable but significant in some patients (grade 4 sepsis and grade 4 gas gangrene). In addition, signs associated with germination, such as abscess formation and destruction of injected tumor masses, were seen in a subset of patients distributed across all six dosing cohorts, thus establishing proof of a mechanism of C. novyi-NT−induced tumor lysis. It is unclear if the traditional drug development paradigm to test multiple escalating dose levels such as 3+3 design used in our study is useful for development of therapy using replicating bacteria. The administered C. novyi-NT dose may be only one of many variables affecting the incidence and robustness of germination and consequently the efficacy of the treatment and the frequency of its AEs. In fact, one of the patients exhibiting the most robust germination was treated at the lowest dose (1 × 104 spores; ref. 6). Unlike other cancer agents, such as small-molecule drugs, whose circulating levels decline after administration, live biologics can exponentially propagate under optimal conditions (15). C. novyi-NT spores germinate under hypoxic conditions; the rate of germination and speed of growth/propagation depends on the tumor microenvironment, which is determined by factors such as tumor size, histology, and host immune competence (15–17). Consideration should be given to these factors when determining the therapeutic index for C. novyi-NT. In addition, the larger size of the injected tumor appeared to be a prominent determinant of germination and grade ≥3 AEs (both P = 0.006). Of interest, similar observations have been reported in canine studies and could be attributed to the more hypoxic state of large tumors. Similar to those canine studies, in our study, AEs related to C. novyi-NT germination were successfully managed with close patient monitoring and appropriate supportive care such as hydration and analgesia with or without systemic antibiotic therapy.
Our study used only a single injection of C. novyi-NT spores into a single tumor, which helps explain why 50% of the patients subsequently started new cancer therapies in the absence of disease progression. Despite this limitation, we observed early signals of clinical activity, including shrinkage of the injected lesions in 41% of evaluable patients and SD as the best overall response for injected and noninjected lesions in 86% of evaluable patients. Nevertheless, the effect of a single dose of C. novyi-NT, although robust, is likely insufficient to reverse the course of advanced cancers; therefore, future clinical trials offering rationally designed combinations and/or repeat injections are warranted. Also, we noticed that germination was associated with a decrease in tumor density rather than a reduction in tumor size, suggesting that common criteria that use tumor size as a surrogate marker of response might not be ideal in this setting. The tumor types treated most frequently in this study were diverse sarcomas, which are known to harbor hypoxic areas (18). Most patients with sarcoma continue to have limited therapeutic options, and early data with immune checkpoint inhibitors have been lukewarm at best, which underscores the need for the development of new therapies and highlights the potential of C. novyi-NT (19). Owing to safety concerns about possible germination and abscess formation, our study excluded patients in whom the tumor targeted for injection was located in any of the three body cavities; thus, the enrollment of patients with other, more common adult solid tumors was restricted.
Patients with germination showed more robust circulating cytokine responses, which is consistent with the notion that bacterial infection causes acute systemic innate immune responses, thereby providing a direct biological measure of germination. The increased cytokine release may contribute to patients' inflammatory responses, leading to systemic toxicities. C. novyi-NT has been shown to induce potent antitumor immune responses, both innate and acquired, in preclinical models (4). Similar to the effects of oncolytic vaccines, C. novyi-NT−induced lysis may stimulate systemic tumor-specific T-cell responses that are measurable in PBMCs (20). Indeed, findings from our ELISPOT assays suggest that a single injection of C. novyi-NT can plausibly increase tumor-specific T-cell responses even in the absence of clinically detectable germination, and in these cases, subclinical germination could not be ruled out. Analyzing paired pretreatment and posttreatment biopsy specimens, we found that 50% of the injected lesions from patients with germination had increased infiltration of T and other myeloid cells, but we found no such increase in patients without germination. However, small numbers and heterogeneity of enrolled tumor types preclude definite conclusion.
In summary, the findings of this phase I study show that a single injection of C. novyi-NT is feasible and can produce robust local tumor destruction and stimulate inflammatory and tumor-specific immune responses. Local and systemic effects can be adequately managed with current risk mitigation strategies; however, AE profile and the need for multiple days observation in the hospital might complicate the future development. Future trials exploiting multiple dosing of a single tumor, multiple injections across more than one tumor and continued refinement/stratification of the patient population most likely to benefit from bacteriolytic therapy are being considered (4). Furthermore, innovative strategies centered at pharmacodynamic endpoints need to be studied to better understand how to investigate bacteria-based therapies in early-phase clinical development as traditional dose-escalation approaches might not be optimal. In addition, a clinical trial investigating a combination with immune checkpoint inhibitor pembrolizumab to augment antitumor immune responses is underway (NCT03435952).
Authors' Disclosures
F. Janku reports other from Biomed Valley Discoveries during the conduct of the study; other from Novartis, Genentech, Astellas, Agios, Plexxikon, Piqur, Symphogen, Bristol-Myers Squibb, Asana, and Upsher-Smith Laboratories; personal fees and other from Deciphera and Cardiff Oncology; personal fees from Guardant Health, IFM Therapeutics, Synlogic, and Immunomet outside the submitted work. S. Goel reports other from Biomed Valley Discovery during the conduct of the study. D.S. Hong reports research/grant funding from AbbVie, Adaptimmune, Aldi-Norte, Amgen, Astra-Zeneca, Bayer, BMS, Daiichi-Sankyo, Eisai, Fate Therapeutics, Genentech, Genmab, Ignyta, Infinity, Kite, Kyowa, Lilly, LOXO, Merck, MedImmune, Mirati, miRNA, Molecular Templates, Mologen, NCI-CTEP, Novartis, Numab, Pfizer, Seattle Genetics, Takeda, Turning Point Therapeutics, and Verstatem; travel, accommodations, expenses from Bayer, LOXO, miRNA, Genmab, AACR, ASCO, and SITC; consulting or advisory role for Alpha Insights, Acuta, Amgen, Axiom, Adaptimmune, Baxter, Bayer, Boxer Capital, COG, ECOR1, Expert Connect, Genentech, GLG, Group H, Guidepoint, H.C. Wainwright, Infinity, Janssen, Merrimack, Medscape, NTRK Connect, Numab, Pfizer, Prime Oncology, Seattle Genetics, SlingShot, Takeda, Trieza Therapeutics, and WebMD; and has other ownership interests for Molecular Match (advisor), OncoResponse (founder), and Presagia Inc (advisor). S.A. Piha-Paul reports other from Abbvie, Inc., ABM Therapeutics, Inc., Acepodia, Inc., Alkermes, Aminex Therapeutics, Amphivena Therapeutics, Inc., BioMarin Pharmaceutical, Inc., Boehringer Ingelheim, Bristol Myers Squib, Cerulean Pharma, Inc., Chugai Pharmaceutical Co., Ltd., Curis, Inc., Daichi Sanko, Eli Lilly, ENB Therapeutics, Five Prime Therapeutics, Gene Quantum, Genmab A/S, GlaxoSmithKline, Helix BioPharma Corp., Incyte Corp., Jacobio Pharmaceuticals Co., Ltd., Medimmmune, LLC, Medivation, Inc., Merck Sharp & Dohme Corp., Novartis Pharmaceuticals, Pieris Pharmaceuticals, Inc., Pfizer, Principia Biopharma, Inc., Puma biotechnology, Inc., Rapt Therapeutics, Inc., Seattle Genetics, Silverback Therapeutics, Taiho Oncology, Tesaro, Inc., TransThera Bio, and NCI/NIH P30CA016672—core grant (CCSG shared resources) outside the submitted work. S.B. Solomon reports grants from Biomed Valley during the conduct of the study; grants from GE Healthcare and personal fees from XACT Robotics outside the submitted work. G.A. DeCrescenzo reports other from BioMed Valley Discoveries, Inc. during the conduct of the study. A. Collins reports other from BioMed Valley Discoveries, Inc. during the conduct of the study. M. Miller reports other from BioMed Valley Discoveries during the conduct of the study. J.L. Salstrom is employed by a contract research organization and has exposure to data from potentially competing clinical trials, but no specific conflicts were present during the conduct of this study. L. Zhang reports other from BioMed Valley Discoveries during the conduct of the study. S. Saha reports other from during the conduct of the study and outside the submitted work. D. Tung is an employee of Biomed-Valley Discoveries, and has nothing additional to disclose. B. Kreider is an employee of BioMed Valley Discoveries who sponsored and supported the Clinical Trial. M. Varterasian reports personal fees from BioMed Valley Discoveries, Inc. during the conduct of the study; personal fees from Takeda, UNUM Therapeutics, Boston Biomedical, Inc., Infinity Pharma, Molecular Templates, Synlogic Therapeutics, Bryologyx, and Skyline Biosciences outside the submitted work. K. Khazaie reports grants from Mayo Clinic during the conduct of the study. M.M. Gounder reports institutional research grant from BioMed Valley Discoveries, personal fees from Epizyme, Springworks, Karyopharm, Daiichi, Bayer, Amgen, Tracon, Flatiron, Medscape, Physicians Education Resource, Guidepoint, GLG and UpToDate; and grants from the National Cancer Institute, National Institutes of Health (P30CA008748)—core grant (CCSG shared resources and core facility). No disclosures were reported by the other authors.
Authors' Contributions
F. Janku: Conceptualization, data curation, formal analysis, supervision, validation, investigation, methodology, writing-original draft, project administration, writing-review and editing. H.H. Zhang: Conceptualization, data curation, formal analysis, supervision, funding acquisition, validation, investigation, methodology, writing-original draft, project administration, writing-review and editing. A.M. Pezeshki: Data curation, formal analysis, investigation, writing-review and editing. S. Goel: Methodology, writing-review and editing. R. Murthy: Conceptualization, investigation, methodology, writing-review and editing. A. Wang-Gillam: Data curation, investigation, writing-review and editing. D.R. Shepard: Data curation, investigation, writing-review and editing. T. Helgason: Data curation, investigation, writing-review and editing. T. Masters: Data curation, investigation, writing-review and editing. D.S. Hong: Data curation, investigation, writing-review and editing. S.A. Piha-Paul: Data curation, investigation, writing-review and editing. D.D. Karp: Data curation, investigation, writing-review and editing. M. Klang: Data curation, investigation, writing-review and editing. S.Y. Huang: Data curation, investigation, writing-review and editing. D. Sakamuri: Data curation, investigation, writing-review and editing. A. Raina: Data curation, investigation, writing-review and editing. J. Torrisi: Data curation, investigation, writing-review and editing. S.B. Solomon: Data curation, investigation, writing-review and editing. A. Weissfeld: Data curation, investigation, writing-review and editing. E. Trevino: Data curation, investigation, writing-review and editing. G. DeCrescenzo: Data curation, investigation, writing-review and editing. A. Collins: Data curation, project administration, writing-review and editing. M. Miller: Conceptualization, project administration, writing-review and editing. J.L. Salstrom: Data curation, project administration, writing-review and editing. R.L. Korn: Data curation, investigation, writing-review and editing. L. Zhang: Conceptualization, data curation, funding acquisition, project administration. S. Saha: Conceptualization, resources, supervision, funding acquisition, methodology, writing-review and editing. A.A. Leontovich: Data curation, investigation, writing-review and editing. D. Tung: Data curation, investigation, writing-review and editing. B. Kreider: Conceptualization, resources, supervision, funding acquisition, writing-review and editing. M. Varterasian: Conceptualization, supervision, methodology, writing-original draft, project administration, writing-review and editing. K. Khazaie: Conceptualization, data curation, supervision, investigation, writing-original draft, writing-review and editing. M.M. Gounder: Conceptualization, data curation, formal analysis, supervision, validation, investigation, methodology, writing-original draft, project administration, writing-review and editing.
Acknowledgments
The study was funded by BioMed Valley Discoveries Inc.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Footnotes
Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/).
Clin Cancer Res 2021;27:96–106
- Received May 27, 2020.
- Revision received July 24, 2020.
- Accepted October 7, 2020.
- Published first October 12, 2020.
- ©2020 American Association for Cancer Research.
References
Volume 27, Issue 1
