Clinical efficacy of anti-programmed death ligand 1 antibody HFC-L1/c4G12 in dogs with malignant tumors: an exploratory study

Abstract

Cancer in dogs remains a major challenge in modern veterinary medicine. Immunotherapy using immune checkpoint inhibitors (ICIs) is available for various human tumor types, and recent veterinary clinical studies have shown that ICIs are a promising approach for treating canine cancers. A canine chimeric anti-PD-L1 antibody, c4G12 (HFC-L1), has been investigated for canine cancer immunotherapy; however, its clinical benefits have not been well characterized in tumors other than pulmonary metastatic (stage IV) oral malignant melanoma (OMM). To explore the efficacy and safety of HFC-L1, we conducted a clinical study in dogs with stage I–III OMM or other tumor types (n=12). HFC-L1 treatment at a dose of 5 mg/kg every 2 weeks was well tolerated, and no grade 3 or higher treatment-related adverse events were reported. Among the dogs eligible for response evaluation (n=10), a partial response was observed in one dog with squamous cell carcinoma, resulting in an objective response rate of 10%. In addition, in a dog with ceruminous cell carcinoma, clinical evidence of a tumor response was observed in metastatic lung lesions. Together, these results suggest that the HFC-L1 therapy is applicable for the treatment of various tumor types, although its clinical benefits should be further evaluated in clinical studies involving a larger number of dogs with each tumor type.

INTRODUCTION

Cancer is a major challenge in modern veterinary medicine and the leading cause of death in dogs [1, 6, 8]. Canine cancers are typically treated with surgery, radiation therapy, and chemotherapy. However, there remains a strong demand to improve treatment outcomes for various tumor types. Immunotherapy has emerged as the fourth pillar of human cancer treatment, largely owing to the success of immune checkpoint inhibitors (ICIs) [28]. Programmed death-1 (PD-1) is an immune checkpoint (immunoinhibitory) receptor expressed on activated T cells that negatively regulates T cell effector function [10, 23]. A ligand for PD-1, programmed death ligand 1 (PD-L1), is expressed in the tumor microenvironment, particularly on the tumor cells [25]. In human patients with renal cell carcinoma and gastric carcinoma, PD-L1 expression in tumors is associated with shorter survival [26, 29], suggesting that the PD-1/PD-L1 axis inhibits the antitumor immune response and is therefore a druggable target for cancer immunotherapy. Indeed, anti-PD-1 or anti-PD-L1 antibodies that reverse the negative regulation of T cells can improve the antitumor immune response, and numerous clinical studies have shown that these ICIs are beneficial for treating various tumor types in humans [9, 23].

Studies in dogs have shown that PD-L1 is expressed in various tumor types [12,13,14, 24], including oral malignant melanoma (OMM), osteosarcoma, nasal adenocarcinoma, and cutaneous squamous cell carcinoma (SCC), at a high positivity rate [13]. Veterinary clinical studies on ICIs have largely focused on OMM treatment. Cumulative evidence suggests the clinical benefit of ICIs in canine OMM, especially in those with pulmonary metastasis (stage IV) [4, 7, 13, 15]. For example, the canine chimeric anti-PD-L1 antibody c4G12, which blocks canine PD-1/PD-L1 binding and enhances T-cell activation in canine peripheral blood mononuclear cell cultures [15], induced tumor regression in a subset of dogs bearing pulmonary metastatic OMM with an objective response rate of 7.7–14.3% [13, 15]. In addition, a survival benefit was strongly suggested in comparison with standard therapies in a disease-matched historical control group treated at the same hospital. The median survival after the diagnosis of pulmonary metastasis was 143 days in the c4G12-treated group, in contrast to 54 days in the historical control group [13]. However, the clinical benefit of c4G12 in dogs with earlier stages of OMM (stage I, II, or III), as well as other tumor types, remains to be studied further, with only a limited number of cases treated in previous studies, where an objective response was observed in dogs with stage II OMM, undifferentiated sarcoma, nasal adenocarcinoma, or osteosarcoma [11, 15].

To further explore the efficacy and safety of c4G12 (renamed here as HFC-L1) in dogs with malignant tumors other than pulmonary metastatic OMM, a veterinary clinical study was conducted in dogs with stage I, II, or III OMM or other tumor types (n=12). The dogs were treated with 5 mg/kg HFC-L1 every 2 weeks as part of multicenter clinical study of HFC-L1 conducted at Hokkaido University (HU), Azabu University (AU), and Obihiro University of Agriculture and Veterinary Medicine (OUAVM).

MATERIALS AND METHODS

Overview of the clinical study

The clinical study was conducted at the veterinary teaching hospitals of AU and OUAVM between November 2021 and February 2024, as part of the multicenter clinical study of HFC-L1, with the approval of the Ethics Committee or Institutional Animal Care Committee (approval numbers: HU, 2021-08; AU, 19403-5 and 220308-1; OUAVM, 21-164). Dogs met the following inclusion criteria for enrolment in the clinical study: (1) dogs diagnosed with stage I, II, or III OMM, as defined by the TNM-based World Health Organization staging scheme [20], or those diagnosed with other tumor types by histopathological or cytopathological assessment; and (2) dogs for whom written informed consent was obtained from the owners. Dogs that met at least one of the following exclusion criteria were excluded from the study: (1) dogs with severe systemic illnesses unrelated to the tumor; (2) dogs that had experienced severe immune-related disorders that might recur during the study; (3) dogs with difficulty in hospital revisits and follow-up observation because of planned relocation or hospital transfer; (4) dogs with any difficulty adhering to the scheduled revisit for drug administration and clinical examinations; (5) Dogs that were (potentially) pregnant or lactating; (6) dogs that had received another experimental therapy within 12 weeks prior to enrolment or were participating in another clinical study at the time of enrolment; or (7) dogs considered by the investigators to be unsuitable for participation in the clinical study. HFC-L1 was prepared as a 10 mg/mL antibody solution dissolved in phosphate-buffered saline (FUSO Pharmaceutical Industries, Ltd., Osaka, Japan) and stored below −20°C until use. The maximum treatment duration specified in the study protocol was up to 96 weeks. When formalin-fixed, paraffin-embedded tumor biopsies (obtained at any time point) were available for immunohistochemistry, PD-L1 expression in tumor cells was evaluated at a commercial pathology laboratory (North Lab, Sapporo, Japan) [13]. HFC-L1 (5 mg/kg) was diluted in saline and administered intravenously over 1 hr using a syringe pump every 2 weeks. Prior to administration, the use of antihistamine drugs was allowed as premedication.

Assessment of safety

Physical examination, complete blood count, and blood chemistry were routinely performed to monitor adverse events. Blood tests were scheduled every 2 weeks for the first 6 weeks and every 6 weeks thereafter. Classification and grading of adverse events were performed according to the Veterinary Cooperative Oncology Group–Common Terminology Criteria for Adverse Events (VCOG-CTCAE) v1.1 [27]. For each adverse event, attribution (causality) was categorized as related, unrelated, or indeterminate by the veterinarians. Treatment-related adverse events (TRAEs) were defined as adverse events that were possibly (related or indeterminate) related to HFC-L1 therapy.

Evaluation of efficacy

The tumor size was measured at least every 6 weeks by clinical examination, thoracic radiography, or computed tomography (CT) using the same modality as the baseline assessment. Ten dogs had at least one measurable target lesion at baseline (i.e., ≥10 mm on clinical examination or CT; ≥20 mm on thoracic radiograph) and were considered “with target disease”. The remaining two dogs had only non-measurable lesions at baseline (“with non-target disease”) and thus were excluded from the response evaluation. The tumor response to HFC-L1 treatment was defined according to the response evaluation criteria for solid tumours in dogs (cRECIST) v1.0 [17]. Complete response (CR) was recorded if all detectable tumors disappeared in response to the treatment, partial response (PR) if the tumor burden was reduced by ≥30%, and progressive disease (PD) if the tumor burden increased by ≥20% or new lesion(s) appeared. To be considered stable disease (SD), the tumor burden must have remained unchanged (decreased by <30% or increased by <20%) for at least 6 weeks. Tumor response was reported as not evaluable (NE) when re-evaluation of the tumor burden could not be performed.

RESULTS

Baseline characteristics of the treated dogs

In total, 12 dogs were enrolled in the study, including nine dogs treated at AU and three dogs treated at OUAVM. Various canine breeds were included, with a median age at the time of enrollment of 12.5 years (range: 6–16). There were nine males (four intact and five castrated) and three females (all spayed). Four dogs had OMM (one stage II and three stage III). The other eight dogs had one of the following tumor types: digit malignant melanoma (n=2), tonsil SCC (n=3), rhinarium SCC (n=1), hepatocellular carcinoma (n=1), or ceruminous gland carcinoma (n=1). At baseline assessment, two dogs (dog #5 with SCC and dog #9 with ceruminous gland carcinoma) did not have any measurable lesions, as defined by cRECIST [17], and were not eligible for response evaluation according to the criteria. Immunohistochemistry was performed on three tumor specimens, and two (one SCC and one OMM) were positive for PD-L1 expression in tumor cells (Table 1 and Supplementary Table 1). HFC-L1 at a fixed dose of 5 mg/kg was administered every 2 weeks, with a median number of treatments of 4.5 (range: 1–7).

Table 1. Characteristics of dogs (n=12) at baseline of HFC-L1/c4G12 therapy.

Characteristic
Breed―no. (%)
Golden Retriever	1 (8.3)
Saluki	1 (8.3)
Shih Tzu	2 (16.7)
Shetland Sheepdog	1 (8.3)
Toy Poodle	1 (8.3)
Norfolk Terrier	1 (8.3)
Pug	1 (8.3)
French bulldog	1 (8.3)
Miniature Dachshund	2 (16.7)
Mix	1 (8.3)
Age―year
Median	12.5
Range	6–16
Sex―no. (%)
Male, intact	4 (33.3)
Male, castrated	5 (41.7)
Female, intact	0 (0)
Female, spayed	3 (25.0)
Tumor type―no. (%)
Malignant melanoma
Oral	4 (33.3)
Digit	2 (16.7)
Squamous cell carcinoma
Tonsil	3 (25.0)
Rhinarium (nasal cavity)	1 (8.3)
Hepatocellular carcinoma	1 (8.3)
Ceruminous gland carcinoma	1 (8.3)
Measurable lesion (s)―no. (%)
Present	10 (83.3)
Absent	2 (16.7)

Safety of HFC-L1 therapy

TRAEs of any grade were reported in one dog (dog #10, 8.3%). No grade 3 or higher TRAEs were observed during the study period (Table 2). Dog #10 developed anorexia (grade 2) and decreased general performance (grade 2) after the first dose of HFC-L1. These conditions persisted during treatment while left untreated, and HFC-L1 treatment was discontinued after the third dose (at week 6) at the request of the owner (unacceptable loss of appetite). The dog was transferred to another hospital and was subsequently lost to follow-up. It is unclear whether the TRAEs were resolved thereafter.

Table 2. Treatment-related adverse events (TRAEs) of any grade (n=12).

TRAEs―no. (%)
Anorexia, grade 2	1 (8.3)
Lethargy/fatigue/general performance, grade 2	1 (8.3)

Evidence of clinical efficacy of HFC-L1 therapy

Among the dogs with target disease (n=10), two died before the first evaluation of the tumor (NE). Another dog (#10) dropped out of the study at week 6 with an incomplete evaluation of the tumor (NE). Six dogs had PD within the first 6 weeks of treatment. The remaining dog (#11) experienced PR as its best overall response (BOR), resulting in an objective response rate (ORR) of 10.0% (95% confidence interval: 0.3–44.5%) (Table 3 and Supplementary Table 1). Dog #11 initially presented with nasal SCC originating in the rhinarium. The dog was treated with palliative radiation therapy (four fractions at 1-week intervals, 6.5 Gy per fraction); however, the tumor progressed gradually throughout the radiation course. Nine days after the last radiation dose, after confirming the tumor progression, HFC-L1 therapy was initiated. The longest diameter of the rhinarium SCC (40 mm at baseline) decreased by 37.5% at week 7 (25 mm, PR) (Fig. 1). At week 13, tumor progression (40 mm, PD) was observed, with a progression-free survival of 93 days. The dog discontinued HFC-L1 treatment at week 15 because of the disease progression.

Table 3. Tumor response to HFC-L1/c4G12 therapy in dogs with target disease (n=10).

Best overall response―no. (%)
CR	0
PR	1 (10.0)
SD	0
PD	6 (60.0)
NE	3 (30.0)
ORR―% (95% CI)	10.0 (0.3–44.5)

CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease; NE, not evaluable. ORR, objective response rate (CR + PR); CI, confidence interval.

Fig. 1.

Antitumor efficacy of HFC-L1 in a dog with squamous cell carcinoma (SCC). The measurable SCC lesion (40 mm at baseline) in the rhinarium responded to HFC-L1 therapy at week 7 (25 mm, partial response).

Among the remaining two dogs with non-target disease, one dog (#5) was treated with HFC-L1 as maintenance therapy after radiation therapy, which was initiated nine days after the last radiation dose without confirmation of tumor progression. Thus, tumor response evaluation was not applicable to this dog. Another dog with non-target disease (#9) had multiple metastatic lesions of ceruminous gland carcinoma in the lungs that developed after surgical resection and adjuvant radiation therapy, as depicted on thoracic radiography (all lesions <20 mm). At week 7 of HFC-L1 treatment, the pulmonary metastatic lesions increased in size and number; however, at week 13, most metastatic lesions disappeared or decreased in size, resulting in a reduction in the overall tumor mass (Fig. 2). Treatment was discontinued at this point after completion of the predefined study period. Figure 3 summarizes the treatment duration and response for each dog (for details, see Supplementary Table 1).

Fig. 2.

Antitumor efficacy of HFC-L1 in a dog with ceruminous gland carcinoma. Non-measurable metastatic lesions of ceruminous gland carcinoma in the lung responded to HFC-L1 therapy at week 13, after transient tumor progression at week 7. Upper panel: Disappearance of the lesion. Lower panel: Shrinkage of the lesion.

Fig. 3.

Treatment duration and response. The duration of HFC-L1 treatment (days) and the timing of key events are shown for each dog. NE, not evaluable; PD, progressive disease; PR, partial response as defined by cRECIST. Note that dogs #5 and #9 did not have any target lesions at baseline.

DISCUSSION

In this exploratory clinical study of HFC-L1, TRAEs were found in only one dog, suggesting that HFC-L1 therapy was well tolerated at the tested dose. An objective response was observed in a dog with rhinarium SCC, with an ORR of 10%. In addition, evidence of clinical efficacy was observed in a dog with ceruminous gland carcinoma, implying that HFC-L1 therapy may be applicable for treating these tumor types. Although the clinical benefits should be verified separately for each tumor type, the results of this study, together with those of previous studies [11, 15], encourage further development of HFC-L1 for the treatment of canine tumors other than pulmonary metastatic OMM.

In this study, owing to the limited availability of tumor specimens, PD-L1 immunohistochemistry was performed in only three cases and two of which were PD-L1–positive. The remaining case of rhinarium SCC (dog #11), which experienced PR as the BOR, was PD-L1–negative in the biopsy sample. Unfortunately, the biopsy was performed before radiation therapy, and tumor sampling immediately before the initiation of HFC-L1 therapy was not feasible, making it unclear whether the target molecule was expressed in the tumor microenvironment during anti-PD-L1 therapy. It is possible that PD-L1 expression might have been induced by the prior radiation, either directly or in response to subsequent inflammatory stimuli [18, 21]. Nevertheless, because immunogenic cell death induced by radiation might be beneficial in enhancing the antitumor immunity reinvigorated by ICIs [4, 21], such a sequential treatment would be a promising approach for treating canine SCC. Although skin SCC expressed PD-L1 in most archival specimens (18/20, 90%) [13], the positivity rate of tonsillar or rhinarium (nasal) SCC remains to be elucidated. As favorable clinical benefits have been suggested for human cutaneous and sinonasal SCC [16, 19], the use of ICIs for this tumor type should be tested in future veterinary clinical studies. To maximize the number of dogs enrolled, tumor sampling for PD-L1 immunohistochemistry was not incorporated in this clinical study due to the invasive nature of the biopsy procedure. Given that PD-L1 expression is inducible in tumor tissues, future studies may need to include a tissue biopsy right before the initiation of anti-PD-L1 therapy to explore the clinical utility of PD-L1 immunohistochemistry as a biomarker of treatment response.

Although dog #9 (with metastatic ceruminous gland carcinoma) was not eligible for response evaluation, as defined by cRECIST [17], the tumor response could be monitored clinically through chest radiography. At week 7, after the third dose of HFC-L1, the tumor progressed, with the emergence (visualization via radiograph) of new lesions; however, during the next three doses, the tumor responded clearly to HFC-L1 treatment. This is consistent with a phenomenon known as pseudoprogression, reported in human immunotherapy studies [3]. Unlike conventional cytotoxic chemotherapy, ICI-induced tumor cell killing is an indirect process that requires a series of steps to establish effective antitumor T-cell responses [2]. An initial apparent increase in tumor size can occur either because of infiltration of immune cells into the tumor lesion or sustained tumor growth until the development of a sufficient antitumor immune response [5]. Such immune-related patterns of response are not anticipated in cRECIST; thus, in future studies on canine ICIs, the response evaluation criteria may need to be modified, as proposed for human clinical studies involving immunotherapy (e.g., iRECIST) [22].

Only a few TRAEs were reported in this study, and no novel safety concerns were identified, although the number of treated dogs was small (n=12). In previous studies using c4G12, some immune-related adverse events were suggested, involving various organs (pneumonia, colitis, hepatitis, pancreatitis, and thrombocytopenia) [4, 11, 13, 15]. The type, frequency, and severity of TRAEs may depend on the characteristics of the study population (including tumor type, breed, age, sex, and prior or concomitant therapy), and safety profiles should be evaluated in future studies involving dog populations in which HFC-L1 is intended to be used.

In conclusion, HFC-L1 therapy at the tested dose was well tolerated and exhibited evidence of clinical efficacy in dogs with SCC or ceruminous gland carcinoma. These results are exploratory but have significant implications for the use of anti-PD-L1 therapy in dogs. Further studies are warranted to identify specific tumor types that are more likely to benefit from HFC-L1 and to maximize its clinical benefits in canine cancer immunotherapy.

CONFLICT OF INTEREST

The test drug (HFC-L1) was manufactured, quality-controlled, and provided for the clinical study by FUSO Pharmaceutical Industries, Ltd. S. Kanazawa, T.A., T.F. and K.Y. are employed by FUSO Pharmaceutical Industries, Ltd. and were involved in drug preparation. They did not play any additional roles in the study design, data collection and analysis, decision to publish, or manuscript preparation. N.M., S. Konnai, Y. Kagawa, S.T., Y.S., T.O., S.M., and K.O. are the authors of patent applications for the materials and techniques described in this paper (PCT/JP2017/029055, PCT/JP2018/011895). The remaining authors declare no competing interests.