Pilot study investigating the intravenous administration of monomeric L-asparaginase to dogs with multicentric lymphoma

Abstract

L-Asparaginase (ASNase) exerts its main anticancer activity by depleting L-Asparagine (Asn). In dogs with lymphomas, ASNase is commonly administered intramuscularly or subcutaneously and incorporated in multiagent protocols. The goal of this study is to assess plasma asparagine depletion and ASNase activity after intravenous administration of ASNase. A secondary goal was to investigate the toxicity and clinical response after intravenous infusion of high-dose ASNase. The study included ten dogs with naïve multicentric lymphoma. All dogs were immunophenotyped, staged, and followed to death. ASNase was administered intravenously at a dose of 3000 IU/kg on a Monday-Wednesday-Friday schedule. ASNase activity and Asn concentration were assessed on days 1, 3, and 5, before and after infusion, and on day 21. Clinical response and toxicity were assessed on weeks 2, 6, and 10, and then every 4 weeks until relapse. Plasma ASNase activity was sufficient to induce and maintain Asn depletion after and between each administration. The activity at day 21 was below detection in 8/10 dogs, and Asn was still depleted in 4/10 dogs. The ORR was 60% (respectively 40%CR and 20%PR). The median TTP was 31 days (95% CI 0–70). All dogs underwent rescue treatment with a multiagent protocol (8 CHOP and 2 LOPP). The median TTP after starting a rescue was not reached after 392 days. Mild gastrointestinal toxicity occurred in 30% of dogs, while hematologic toxicity or hypersensitivity reactions did not occur. ASNase activity and depletion of Asn were clinically adequate. The clinical efficacy was comparable with previous literature. The intravenous infusion of high-dose ASNase was well tolerated with only mild and self-limiting toxicity.

Introduction

Asparagine is a non-essential amino acid in normal cells but is considered essential for leukemic and lymphoid cells because they have a reduced ability to synthesize asparagine themselves and rely on exogenous sources for survival [1, 2]. L-Asparaginase (ASNase) is an enzyme that selectively kills leukemic and lymphoid cells by depleting the extracellular sources of asparagine (Asn) [1,2,3].

ASNase was originally purified from E. coli, but since then many different forms of native, pegylated, or recombinant ASNase have been developed to reduce its immunogenicity, which seems to be the limiting toxicity [1, 2, 4, 5]. In humans, ASNase can be administered intravenously, intramuscularly, or less commonly subcutaneously, while in veterinary medicine only the intramuscular, subcutaneous, and intraperitoneal routes are currently described [1,2,3, 6,7,8,9].

In veterinary medicine, intramuscular or subcutaneous ASNase is used in the treatment of canine lymphoma as single-agent or in the induction phase of multiagent protocols. The reported response rates are variable and range from 30 to 82% for both PEG-ASNase and native ASNase [7, 8, 10].

The efficacy and toxicity of native or pegylated ASNase administered as an intravenous infusion in dogs with lymphoma has never been evaluated.

In-vitro studies suggest that the anti-tumor effects of ASNase might not be exerted through asparagine depletion alone, but are also mediated by direct cytotoxicity [11].

Several clinical studies have shown that the efficacy of ASNase is dose-dependent, which indicates that not only asparagine depletion but also the enzyme molecules play a direct role in the killing of cancerous cells. Since only a fraction of lymphoma cells are in the blood stream at any time, we have decided to include interstitial fluid in the potential distribution volume. Therefore it might be possible that the efficacy of ASNase could be increased if the enzyme was available closer to the target cancer cells, in the interstitial fluid.

In this study a dose of 3000 IU/kg was chosen assuming that extravasation of monomeric asparaginase will be much higher than of a standard tetrameric form in the hope that the concentration in the interstitial fluid will reach the level of about 1 IU/ml which, from in vitro work, is needed to get substantial kill of lymphoma cells. It is suppsed that the conventional 400 IU/kg does not get close to that.

Extravasation of ASNase can be increased by reversibly dissociating its tetrameric form (of about 140 kDa) into monomers (of about 35 kDa). This can be achieved by infusing ASNase dissolved in urea (5 molar) [12]. Urea is quickly diluted in the blood allowing for reconstitution of tetrameric, active form of ASNase both intra- and extra-vascularly. Furthermore, concurrent administration of insulin (as insulin/glucose clamp), increases the permeability of the capillary bed for both monomeric and tetrameric forms of ASNase [13]. Insulin is also a potent growth factor and as such it inhibits protein breakdown and stimulates protein synthesis, aiding the depletion of all amino acids, including Asn.

The primary aim of this study was to measure the plasma concentration of Asn and the plasma activity of ASNase after intravenous administration of ASNase dissolved in urea and administered with insulin and glucose in dogs with lymphoma. The secondary objective was to report toxicity and clinical response to the intravenous administration of the new formulation of ASNase.

Materials and methods

Patient selection and evaluation

In the period between September 2022 and June 2023, ten dogs were enrolled for the pilot study at the AniCura Animal Oncology and Imaging Center, Hünenberg, Switzerland. Criteria for enrollment were the diagnosis of multicentric large cell lymphoma (cytology or histopathology). Dogs who had previously received chemotherapy, corticosteroids, or those with a history of pancreatitis and/or cardiac and/or renal diseases, or those without owner consent were excluded. This study was approved by the cantonal ethical authority of Canton Zug in Switzerland (approval number: 35089; ZG/118/2022).

We collected data on age, sex and neutering status, breed, weight, diet, clinical stage and substage, immunophenotype, starting date of the treatment, treatment response, time to progression, rescue chemotherapy protocol, and time to progression to the first rescue protocol. The World Health Organization (WHO) staging system was used to categorize the clinical stage and substage [14]. The Veterinary Cooperative Oncology Group (VCOG) criteria were used to assess the response to the treatment and the toxicity [15, 16].

Diagnosis and staging

Large cell multicentric lymphoma was diagnosed with cytology or histopathology. The immunophenotype was determined by flow cytometry or immunocytochemistry, on samples obtained from the lymph nodes. All dogs underwent staging with hematology, blood film evaluation, serum biochemistry, urinalysis, thoracic radiographs, abdominal ultrasound, and bone marrow aspirate (6/10). The peripheral blood smears were all examined by a certified clinical pathologist.

Treatment protocol

All ten dogs underwent the following treatment protocol: constant rate infusion of Insulin Lispro (Humalog, Eli Lilly, 0.5 IU/kg/h IV), constant rate infusion of glucose (Glucose 20%, B. Braun, 1 g/kg/h IV), constant rate infusion of ASNase (ASPAREA®; Hepius Biotech AG; 3000 IU/kg) administered over 6 h on a Monday-Wednesday-Friday schedule in one week. Before infusion, clemastine (Tavegyl, Gebro Pharma, 2 mg/2 ml, 0.05 mg/kg IV) and maropitant (Cerenia, Zoetis 1 mg/ml, 1 mg/kg IV) were administered.

Five dogs received one dose of ASNase (Leunase, 10,000 IU, or 400 IU/kg IM) one week before starting the protocol in order to ease the symptoms of the lymphoma until ASNase infusion protocol could be performed. At the time of the treatment, one dog was receiving Carprofen (RIMADYL®, Zoetis, 100 mg tablets, 3.4 mg/kg q24h) for the management of chronic joint disease.

The follow-up evaluation was done at 2 weeks after the last administration of ASNase and then every 4 weeks for the first 14 weeks or until relapse if it occurred before 14 weeks. If the dog was still in complete remission restaging including hematology, biochemistry, thoracic radiographs, and abdominal ultrasound was repeated after 14 weeks and then every 12 weeks.

Dogs with documented progressive disease were promptly started on CHOP, if they were diagnosed with B-cell lymphoma, or LOPP, if they were diagnosed with T-cell lymphoma.

Response assessment

The therapeutic response was determined using the Veterinary Cooperative Oncology Group response evaluation criteria [15]. A complete response (CR) was characterized by the disappearance of all measurable diseases; a partial response (PR) was characterized by a decrease (> 30%) in the sum of longest diameters of target lesions; stable disease (SD) was characterized by a < 30% decrease or < 20% increase in target lesions compared to baseline; and progressive disease (PD) was characterized by a > 20% increase in target lesions or the development of a new lesion. The overall response rate is calculated adding the percentage of dogs in CR and in PR. Dogs with SD or PD will be considered non-responders.

Toxicity

The major side effects of ASNase administration are hypersensitivity reactions, gastrointestinal toxicity, and pancreatitis. In our study, gastrointestinal side effects were categorized according to the Veterinary Cooperative Oncology Group-Common Terminology Criteria for Adverse Events [16]. Myelotoxicity was assessed through hematology and blood smear evaluation 2 weeks after the last administration of ASNase.

L-asparagine and L-asparaginase activity in frozen plasma

In all 10 dogs, L-asparaginase activity was evaluated immediately before and after every administration of ASNase, and then 2 weeks after the last administration (on day 21). The blood was collected in syringes chilled in iced water, immediately processed to obtain EDTA plasma, and then frozen for analysis.

The plasma asparagine concentration was measured by Hepius Biotech (Zürich, Switzerland) on an amino acid analyzer (LA8080 High Speed Amino Analyzer, Hitachi, Japan). ASNase activity was measured by Next Molecular Analytics (Chester, VA, USA). Activity is detected in a coupled enzyme assay. In the first reaction, L-asparagine is converted to L-aspartate and ammonia in the presence of ASNase. In the second reaction, L-aspartate and a-ketoglutarate are converted into oxaloacetate and L-glutamate by glutamate oxaloacetate transaminase. In the final reaction, oxaloacetate and NADH are converted to malate and NAD + by malate dehydrogenase. The reaction progress is reflected by the decrease in NADH concentration which is monitored by absorbance at 340 nm. The assay was customized to the specific ASNase used in this study.

In human patients, the ASNase trough activity in plasma is considered adequate if ≥ 0.4 IU/ml or even ≥ 0.1 IU/ml [17].

Statistical analysis

The small number of patients allowed only a minimal statistical analysis. The study ended on May 31^st, 2024, and all dogs alive on this date were censored. We calculated time-to-progression (TTP from the day of the first administration of ASNase until relapse); Rescue time-to-progression (Rescue-TPP) from the day of starting rescue chemotherapy protocol until relapse, death, or the end of the study; overall survival (OS) from the day of first administration of ASNase until death or the end of the study. A KM product-limit method was used to estimate TTP, Rescue-TTP, and OS with a 95% Confidence Interval. Statistical analysis was performed using commercial software (IBM SPSS Statistics, version 23.0).

Results

Patient population

In this study, we enrolled 10 dogs with a diagnosis of large cell multicentric lymphoma, including 4 neutered females, and 6 males (2 intact, 4 castrated). The breeds were Beagle (n = 1), Dogo argentino (n = 1), Jagdterrier (n = 1), Miniature schnauzer (n = 1), Shiba-Inu (n = 1), Yorkshire terrier (n = 1), mixed breed (n = 4). The median body weight was 17.4 kg (range, 4.8 – 40.6 kg), and the median age was 8.75 years (range, 6—12 years).

Diagnosis and staging

All patients had at least four enlarged lymph nodes at presentation. The most common symptoms at diagnosis, apart from lymphadenomegaly, were hyporexia (n = 3), lethargy (n = 2), gagging (n = 2), diarrhea (n = 1), erythema (n = 1), hematochezia (n = 1), hyperthermia (n = 1), panting (n = 1), vomiting (n = 1).

Complete blood count examination, serum biochemistry, and urinalysis were performed in all dogs at presentation. The abnormalities observed in the complete blood count were lymphopenia (n = 4), monocytosis (n = 4), thrombocytopenia (n = 4), anemia (n = 2), lymphocytosis (n = 2), leukopenia (n = 1), neutrophilia (n = 1), reticulocytosis (n = 1). The abnormalities observed in the serum biochemistry were alanine aminotransferase (ALT) elevation (n = 2), alkaline phosphatase (ALP) elevation (n = 2), gamma-glutamyl transferase (GGT) elevation (n = 1), hypoglycemia (n = 1). The abnormalities observed in the urinalysis were proteinuria (n = 3) and crystaluria (n = 2).

In all dogs, thoracic radiographs and abdominal ultrasound were performed. The thoracic radiographs showed the following abnormalities: sternal lymphadenomegaly (n = 2), cranial mediastinal lymphadenomegaly (n = 2), tracheobronchial lymphadenomegaly (n = 1). In the abdominal ultrasound the main alterations were splenopathy (n = 8), hepatopathy (n = 6), abdominal lymphadenomegaly (n = 10), testicular nodule (n = 1), pancretopathy (n = 1), cystopathy (n = 1), prostatopathy (n = 1), cholelithiasis (n = 1).

Cytology of the spleen and liver was performed in all dogs but two (due to severe thrombocytopenia). Concurrent infiltration of liver and spleen was the most common finding (n = 7), followed by only spleen infiltration (n = 1). Bone marrow aspirate was performed in 6 dogs, one sample was positive for lymphoma infiltration, four samples were negative for lymphoma infiltration, and one sample was not diagnostic. In the other cases, the bone marrow aspirate was not performed because in 3 cases the lymphoma was classified as stage V based on the blood smear or skin involvement, and in one case the owner did not give consent. There were four stage V, five stage IV, and one stage III lymphomas. Six dogs were substage a, and four dogs were substage b. The immunophenotype was determined by the flow cytometry in 9 dogs, and by immunocytochemistry in 1 dog. We documented eight B-cell and two T-cell lymphoma immunophenotypes.

Treatment response

The treatment response was evaluated in all 10 dogs two weeks after the completion of the ASNase course. The response rate was 6/10 (60%), with four dogs in complete remission (CR) and two dogs in partial remission (PR). Stable disease (SD) was observed in 3/10 dogs. Progressive disease (PD) was documented in 1/10 dogs. At two weeks, the clinical symptoms observed at the time of diagnosis were resolved in all dogs.

Outcome

The median follow-up time was 382 days (range 92 – 571). For the six dogs that achieved complete or partial response, mean and median TTP was respectively 107 days (95% CI 6 – 208) and 31 days (95% CI 0 – 70) (Fig. 1). Only one dog was censored from the TTP analysis because it was still in complete remission at the time of writing after 571 days. Nine dogs, the 4 non-responders and the 5 dogs with PD after ASNase had chemotherapy; seven B-cell lymphomas had CHOP and two T-cell lymphomas had LOPP. Four dogs had a relapse after starting rescue protocol after a median of 98 days (range 48 – 350). The remaining 5 dogs were still in remission by the end of the follow-up time. (range 277 – 458) (Fig. 2).

Fig. 1

Kaplan–Meier curve indicating the time to progression (TTP) after L-asparaginase administration, expressed in days

Fig. 2

Kaplan–Meier curve indicating the time to progression after starting the rescue protocol (rescue TTP), expressed in days

Four dogs died during the study period for causes related to the lymphoma and 6 were still alive. The mean OS for the 10 dogs was 297 (95% CI 179 – 416), while the median OS was not reached (Fig. 3).

Fig. 3

Kaplan–Meier curve indicating overall survival, expressed in days

Adverse events

No hypersensitivity reactions were observed during or after the intravenous administration of ASNase. None of the dogs developed hematological toxicity at week 2. Three dogs developed GI toxicity: 1 dog Grade I diarrhea on day 3, 1 dog Grade II anorexia on day 3, and 1 dog grade II vomiting on day 4.

No major changes related to blood glucose parameters were observed.

L-asparagine and L-asparaginase activity

In 9 dogs 7 samples were collected before and after each administration of ASNase and then 2 weeks later. In 1 dog the sample before the first administration was not collected. Asparagine was depleted to below detection (1 μmol/l) in all dogs after the first ASNase administration and was still depleted on day 21 in 40% (4/10) of dogs (Fig. 4). ASNase activity was adequate in all ten dogs after each administration for at least 2 days. On day 21 two dogs had activity of 0.15 and 0.26 IU/ml while in 8 dogs activity was below the detection limit (0.013 IU/ml) (Fig. 5). The mean ASNase half-life in plasma was 20.7 h with a standard deviation of 5.5 h (two-point calculation over 42 h for 20 infusion sessions from Mon-Wed and Wed-Fri).

Fig. 4

Chart showing the plasmatic L-asparagine level, expressed in µmol/L, before and after each day of administration, and on day 21 after the first administration. Monday before the administration (Mon before), Monday after the administration (Mon after), Wednesday before the administration (Wed before), Wednesday after the administration (Wed after), Friday before the administration (Fri before), and Friday after the administration (Fri after). Each dot represents a patient, and the line represents the mean value. Dog 1, 3, 7, 8, 9 received ASNase one week prior ASPAREA®

Fig. 5

Chart showing the plasmatic L-asparaginase activity level, expressed in IU/mL, before and after each day of administration, and on day 21 after the first administration. Monday before the administration (Mon before), Monday after the administration (Mon after), Wednesday before the administration (Wed before), Wednesday after the administration (Wed after), Friday before the administration (Fri before), and Friday after the administration (Fri after). Each dot represents a patient, and the line represents the mean value. Dog 1, 3, 7, 8, 9 received ASNase one week prior ASPAREA®

Discussion

The primary aim of the present study was to assess plasma asparagine depletion and ASNase activity after intravenous administration of ASNase dissolved in urea augmented with insulin/glucose clamp in dogs with lymphoma. Asparagine was depleted to below detection (1 μmol/l) at all sampling times from immediately after the first infusion and it remained undetectable for at least 42 h after consecutive infusions. On day 21, two weeks after the last infusion, the plasma concentration of asparagine was still below detection in 40% of the cases (4/10). The degree of asparagine depletion was not investigated in a shorter period, but it is fair to suppose that it might be higher than 40%.

In human medicine it is not known a precise time of depletion needed to obtain leukemic cells killing, nevertheless it is shown that the longer the depletion the better the tumor control.

The conventional schedule of Mon-Wed-Fri accepted in human oncology decades ago is thus confirmed as adequate in dogs [18, 19]. As an induction in treating leukemia in human medicine, the Mon-Wed-Fri schedule is repeated for 10 to 11 consecutive weeks with naked E. coli asparaginase. Pegylated versions have reduced the frequency of application by about factor 5 to 6, but the overall period is still about the same. In current veterinary use ASNase is typically delivered in a single session followed by one of the multi-drug protocols, e.g., CHOP. It is thus not surprising that the veterinary literature provides little to no support for the use of ASNase in combined protocols. The prevailing recent approach is to use a single injection of ASNase followed by chemotherapy in rescue treatments [20, 21].

In our study plasma activity at a 42-h trough was an order of magnitude higher than what is considered by consensus as adequate in the use of ASNase in human oncology.

In human medicine, Asn depletion assessment in plasma is not a good index of the treatment success, so the measurement of ASNase activity level is usually preferred, to monitor the efficacy of ASNase treatment in patients with acute lymphocytic leukemia. The main reason is the imprecise assessment of Asn depletion, because the enzyme ASNase keeps hydrolyzing plasma Asn also after blood sampling, leading to a falsely decreased Asn blood levels. In addition, measurement of ASNase activity is helpful in detecting silent inactivation, where patients have immunogenic inactivation of ASNase without showing symptoms (hypersensitivity, pancreatitis, etc.) [22].

Most anti-cancer drugs, including ASNase, are discovered and tested in vitro to determine the effective concentration levels necessary for efficacy. In these tests, the key variable is the concentration/activity of the enzyme. In most of the tested lymphomatous/leukemic cell lines, significant cell killing is documented at approximately 1 IU/ml of ASNase concentration/activity, but asparagine can be depleted from the culture medium at a concentration/activity of ASNase as low as 0.1 IU/ml. The conclusion is that the depletion of asparagine should be considered a necessary but not sufficient condition for lymphomatous/leukemic cells killing.

These preliminary in vitro studies were translated into clinical practice and asparaginase-enzyme activity levels are used in human oncology to monitor patients receiving asparaginase therapy. The threshold accepted to determine efficacy was derived from this data and set at 0.1 IU/ml of ASNase activity. The service of measuring ASNase activity offered by Next Molecular Analytics (www.nextmolecular.com) is used by most oncology clinics in the USA (and some internationally). However, no sample from a dog treated by ASNase has ever been received by Next Molecular until those from the present study.

A bolus infusion of ASNase at 3000 IU/kg, assuming 40 ml/kg of plasma volume, would result in 75 IU/ml plasma activity. At the end of 6-h infusions, the average plasma activity in this study was about half of that. With the concentration of monomers infused and the time allowed, intravascular reconstitution of active tetramers is mostly completed. The missing 50% of activity was thus either lost due to physiological inactivation, or it was relocated to extravascular fluids. Future research will hopefully provide some clarification. The data collected in the present study shows enzymatic activity at trough (about 8 IU/ml, at 42 h) to be at least an order of magnitude higher than what the current human oncology literature suggests as adequate. Admittedly, the clinical outcomes, especially in childhood ALL, collected over decades in thousands of patients, provide strong evidence that ASNase is effective. However, one can still ask whether the efficacy could be improved. The high dose used in the present study was chosen to account for expected restricted extravasation in the hope of still getting the activity in the interstitial fluid to a relevant level. There is ample human clinical literature suggesting that the higher dosing of ASNase leads to improved outcomes.

The design of the study does not allow for a rational comparison to prior use of ASNase in dog lymphoma. The benchmark study of MacEwen et al. [10] was the first published study to compare naked E. coli ASNase to a pegylated version. Two injections a week apart were followed by a combined chemo-protocol a week after the second injection. There was no difference between the naked and pegylated version. The response criteria used at that time were different from what we have used in the present study. Perhaps the only comparable response criterium is a complete response (CR). In the PEG cohort, they reported 21% CR, and in the naked cohort 23% CR. Statistical power with 10 dogs in our study and 50 in theirs is too low for a meaningful comparison but 40% CR in our study looks promising. In addition, among the 5 dogs receiving ASNase before the intravenous protocol with ASPAREA® just two dogs showed a longer response to the treatment alone, otherwise the remaining three did not show a better outcome. Specifically two of these five dogs had progressive disease already at the first 2 weeks re-check after completion of the ASNase administration and they also had the shortest overall survival. Therefore, it is thought that the pre-treatment with additional intramuscular ASNase did not improve the response to the treatment.

None of the patients developed a hypersensitivity reaction or hematological toxicity and there was no evidence of silent inactivation. The incidence of allergic reactions reported in other studies is 4% and therefore it is likely that the number of dogs was too low to detect this side effect [10, 23]. Three dogs (30%) developed mild (grade I or grade II) gastrointestinal adverse effects.

Conclusions

This study showed that ASNase dissolved in 5 molar urea was safely administered intravenously, even at a high dose, after pre-administration of antihistamines in these 10 dogs. None of the patients developed a hypersensitivity reaction or hematological toxicity and there was no evidence of silent inactivation. The incidence of allergic reactions reported in other studies is 4% and therefore it is likely that the number of dogs was too low to detect this side effect [10, 23]. Three dogs (30%) developed mild (grade I or grade II) gastrointestinal adverse effects.

The use of a 6-h infusion of ASPAREA® augmented with an insulin/glucose clamp is clinically impractical. In the next planned clinical study, a 1-h infusion of ASPAREA® will be tested as a single infusion to be followed a week later by one of the standard chemotherapy protocols.