Human colorectal cancer is one of the most common oncological diseases in the world, characterized by high metastatic activity and frequent complications [1, 2]. According to the WHO, in 2023, colorectal cancer was the third most common oncological disease worldwide, accounting for about 10% of all registered oncopathology cases. It is the second most common cause of cancer-related deaths. Early diagnosis and timely initiation of treatment of colorectal cancer significantly increase the life expectancy of patients and improve the prognosis of treatment. However, the disease is often diagnosed at late stages, which limits the selection of treatment options. Currently, radical surgery is the core treatment approach to colorectal cancer, but, as a rule, it is effective mainly in the early stages of the disease, when the tumor has not yet spread through the body. Other basic treatment methods are radiation therapy, chemotherapy, immunotherapy, and their combinations [3], but their effectiveness is low without proper cytoreduction. These approaches are prioritized at late stages of the disease. It has been established that in colorectal cancer with extensive liver metastases, the five-year survival rate can reach 58% only if the treatment is multimodal and includes neoadjuvant chemotherapy, intraarterial chemoembolization, liver resection, immunotherapy, and radiofrequency ablation; these findings underscore the limited therapeutic potential of any monotherapy [4]. Moreover, drug resistance in tumor cells, driven by intrinsic mechanisms of tumor progression and commonly developing with most current cytostatic agents, is a major cause of the rapid loss of chemotherapy efficacy.

Oncolytic viruses are a promising class of therapeutic agents for malignant neoplasms: they selectively replicate in tumor and tumor-associated cells and exert antitumor effects through direct oncolysis, activation of immune effector mechanisms, and modulation of tumor angiogenesis [5–7]. One such agent is Newcastle disease virus (NDV; Orthoavulavirus 1, family Paramyxoviridae), which can selectively infect tumor cells with impaired antiviral interferon signaling [8]. Genetically engineered NDVs have been extensively studied in animal experiments and clinical trials [9–12]. Natural strains isolated from wild birds also exhibit pronounced oncolytic properties and do not require genetic modification, but their therapeutic potential remains to be evaluated experimentally.

One promising strain is NDV/Altai/pigeon/770/2011, which was isolated in 2011 from a rock pigeon (Columba livia) in the Altai Region of the Russian Federation and classified as mesogenic [13]. This strain serves as the parental strain for candidate oncolytic virotherapy formulations currently under evaluation. The preparation is a cultured virus-containing fluid purified from the globulin-protein fraction by tangential-flow filtration using a 100-kDa molecular-weight cutoff membrane. Hydroxyethyl starch (4%) or glucose (5%) is included as a stabilizer to maintain viral infectivity and infectious titer.

This study aimed to evaluate the antineoplastic activity of laboratory prototypes of an NDV-based antitumor drug and to compare their efficacy with that of standard cytostatic therapy in a murine model of colon adenocarcinoma (AKATOL model; CT26 cell line).

METHODS

We used 60 male BALB/c mice aged 2–2.5 months and weighing 20–22 g, obtained from the Shared Use Center SPF Vivarium of the Institute of Cytology and Genetics, Siberian Branch of the Russian Academy of Sciences. For the first set of the experiment, 35 mice were used and divided into seven groups of five animals each. For the second set, the remaining 25 mice were used and divided into five groups of five animals each. The transplanted tumor model was murine colon adenocarcinoma (AKATOL model; CT26 cell line). This model is widely used in fundamental and preclinical studies and is characterized by specific patterns of tumor growth and metastasis, as well as particular metabolic features, receptor expression profiles, and signaling pathways [14–16].

The tumor model was established using a standard protocol: CT26 cells were implanted subcutaneously into the right hind limb thigh of mice at a dose of 1 × 10⁶ cells per animal in 0.2 mL of isotonic saline.

The experiment consisted of two sets. The first set followed a classical approach: drug administration was initiated 48 hours after tumor cell inoculation. For the second set, we focused on the standards of an antitumor effect assessment experiment; treatment was initiated once a solid tumor nodule had formed and the tumor had reached a predefined size.

The reference drugs were Endoxan (cyclophosphamide; Baxter Oncology GmbH, Germany) and experimental NDV-based solutions, an unpurified virus preparation propagated in VeroE6 cell culture and a virus purified by tangential-flow ultrafiltration, with hydroxyethyl starch or glucose added to preserve infectious titer. The titers of all virus-containing preparations ranged from 10^5.7 to 10^6.0 TCID₅₀/mL (table). In in vitro studies and preliminary laboratory experiments, this dose was found to be the most effective and did not cause pronounced toxic effects in non-target cell populations or body tissues.

Experimental samples of the antitumor drug based on the oncolytic NDV/Altai/pigeon/770/2011 strain were prepared as a solution containing replication-competent infectious viral particles cultured in the Vero E6 continuous cell line. The first experimental sample was unprocessed; it was collected at the end of a 3-5-day culture period. To destroy the Vero E6 cells and release viral particles from the cells and cell surfaces into the culture medium, the collected material was frozen at –20 °C for 24 h. The virus-containing material was then thawed at room temperature, and the accumulated viral particles were passed through a 0.45-µm filter to remove cell debris from the resulting virus-containing fluid. In addition to the virus, the preparation contained amino acids, buffer salts, and other components of the culture medium and serum, as well as certain proteins, peptides, and cellular enzymes.

The other two experimental samples were virus-containing liquids purified by ultrafiltration. To remove fetal bovine serum proteins, peptides, and low-molecular-weight compounds (salts, sugars, amino acids, nucleotides, and buffer components) present in the cell culture medium, the liquid was purified by tangential-flow ultrafiltration using a 100-kDa molecular-weight-cutoff membrane. The system included a VivaFlow® 200 cassette (Sartorius, Germany) and a peristaltic pump. Ultrafiltration yielded purified virus-containing samples, which were subsequently evaluated for antitumor activity. Preservation of infectious titer required long-term storage at –70 °C. In the absence of stabilizers, low-temperature storage and subsequent thawing can disrupt the protein envelopes of enveloped viruses, resulting in loss of infectivity [17, 18]. Therefore, to stabilize the virus in solution, preserve its infectivity, and extend its shelf life, hydroxyethyl starch (HES) or glucose was added to the purified liquid containing the oncolytic NDV strain. Thus, two virus-containing preparations were produced: one stabilized with 3% HES and the other with 5% glucose.

In the first set of the experiment, cyclophosphamide was administered intraperitoneally in 0.9% NaCl solution on the second and fifth days after tumor implantation, and in the second set — when the tumor reached a size of 150–400 mm³. The maximum volume of the administered preparation did not exceed 0.5 ml. The dose of the administered drug was 175 mg per a kilogram of the animal's body weight.

Taking into account the incomplete understanding of the mechanisms underlying antitumor action, we divided the experiments with the virus-containing preparations into two parts: in the first, the drugs were administered IV daily in a volume of 100 µl per animal for five days, starting 48 hours after implantation; in the second, treatment began after the neoplasm reached a size of 150–400 mm³ (the first and the second sets of the experiment, respectively).

In the 1^st group, the mice were implanted tumors and remained untreated (control group). In the 2^nd group, the animals received Endoxane (cyclophosphamide group). Specimen of the 3^rd group were injected with an unpurified form of the virus-containing drug (unpurified group). Mice of the 4^th and 5^th groups were administered purified preparations stabilized with hydroxyethyl starch (HES group) or glucose (glucose group), respectively.

At the first stage of the study, we analyzed the indicators characterizing the experimental tumor process: 1) the duration of the latent period of tumor growth in days, established by measuring the time to registration of primary tumor nodules (PTN) in the malignant cell transplantation area; 2) mean survival time (MST) and life expectancy increase (LEI), both in days, the latter calculated as LEI (%) = [MSTexperiment — MSTcontrol] / MSTcontrol × 100%, with clinically significant level accepted at above 25%; 3) the duration of the tumor growth period from the moment of detection of the nodule to the death of the animal, in days; 4) the volume of PTNs in mm³, calculated as A × B2/2, where A is the major axis of the tumor, B is the minor axis of the tumor [19]; and tumor growth inhibition (TGI), calculated as TGI (%) = [(Vcontrol — Vexperiment) / Vcontrol] × 100%, with clinically significant level accepted at over 50%. After the tumor reached the maximum allowable volume of 2000 mm³, mice were euthanized by high doses of anesthesia (Zoletil, injection); the fact of the animal's death was additionally confirmed by the destruction of the structures of the central nervous system. During the second set of the experiment, we evaluated only the growth dynamics of the neoplasm, and used the TGI formula for this purpose.  

The obtained results were analyzed using nonparametric statistical methods: the Kruskal–Wallis test and Pearson’s chi-square test. The level of statistical significance was set at p < 0.05. The time to PTN registration, the duration of tumor nodule growth from detection to the death of the animal, and the mean survival time of the mice are presented as Me [Q₁; Q₃] (median and interquartile range). Statistical analysis was performed using STATISTICA 8.0 software (StatSoft Inc., Tulsa, OK, USA).

RESULTS

First set of the experiment: investigation of the antitumor potential of drugs when administered before the tumor has formed

The primary aim of the first set of experiments was to evaluate the antitumor efficacy of the virus-containing preparations studied during the latent period of tumor growth. The time to PTN detection was used as an overall indicator of the dynamics of tumor initiation in vivo (fig. 1).

In the early treatment experimental groups tumor nodules appeared later (on average on the seventh day of the study) than in the control group.

Cyclophosphamide increased the latent stage of the tumor progression by an average of eight days compared with the control group, which is significant (p = 0.003). A similar effect was observed in the group that received unpurified drug, but the mean extension of the latent stage was five days (p = 0.017), which is still significant. The differences between the cyclophosphamide and the unpurified groups were not statistically significant (p = 0.281), which indicates a comparable antitumor efficacy of the drug and the preparation against the respective cancer.

In the HES group, the latent period was, on average, three days longer than in the control group; this difference only showed a trend (p = 0.093). A similar result was observed in the glucose group, where the latent period was also prolonged by approximately three days compared with the control group (p = 0.107), suggesting low efficacy of this preparation against the tumor type studied.

Administration of the reference cytostatic drug yielded an increase in the mean survival time by 54.2% (p = 0.073) (fig. 2). In the unpurified group, this indicator increased by 37.5% compared with the control, which is statistically significant (p = 0.041). At the same time, the difference with the cyclophosphamide group was insignificant (p = 0.371).

Experimental therapy with purified CT26 virus-containing preparations supplemented with HES or glucose showed limited efficacy: mean survival time increased by no more than 8.3% in the HES group and by 16.7% in the glucose group.

Assessment of the duration of the tumor growth period from the moment of detection of the nodule to the death of the animal reflects the direct effect of drugs on the tumor.

By this indicator, cyclophosphamide was effective: it prolonged the survival time by an average of 6 days (p = 0.026), which is significant. All virus-containing preparations tested were insufficiently effective, suggesting a relatively weak direct antitumor effect in this animal model.

Tumor growth dynamics were assessed based on changes in the volume of PTNs. For this purpose, tumor dimensions were measured at three time points — days 15, 20, and 25 after subcutaneous adenocarcinoma transplantation.

The reference antitumor agent, administered early after tumor transplantation, effectively inhibited the growth of PTNs in the murine colon adenocarcinoma model. Compared with the control group, we observed a significant reduction in tumor volume (fig. 3): by 77.6% on day 15 (p = 0.020), by 85.2% on day 20 (p = 0.006), and by 83.8% on day 25 (p = 0.053).

The unpurified virus-containing preparation showed colon adenocarcinoma growth inhibition effect comparable to that of the reference drug. Against the control group, the PTNs in the unpurified group were smaller by 74.5% on day 15 (p = 0.037), 81.5% on day 20 (p = 0.094), and 76.3% on day 25 (p = 0.180). There were no statistically significant differences compared to the cyclophosphamide group.

The glucose-stabilized purified form of the test drug temporarily suppressed tumor growth relative to the control group, with inhibition of 62.6% on day 15 and 62.9% on day 20 after tumor transplantation. However, by day 25, the effect of the preparation had nearly disappeared (13.8%), indicating a limited duration of its antitumor activity.

Therapy with the HES-stabilized purified form of the test drug failed to inhibit tumor growth throughout the observation period and had virtually no effect on the measured parameter.

Second set of the experiment: investigation of the antitumor potential of drugs when administered after the tumor has formed

The second set of experiments enabled evaluation of the antitumor activity of the tested drugs against a sufficiently large tumors within their characteristic microenvironment and vascular bed.

In our opinion, this stage of the study is more consistent with the typical clinical picture of cancer, when the tumor is significant enough for visualization upon detection and has characteristic structural features.

Figure fig. 4 presents the results of the analysis of changes in the volume of the tumor in experimental groups.

Cyclophosphamide effectively inhibited tumor growth in murine colon adenocarcinoma model. Its administration yielded a statistically significant reduction in PTN volume, averaging 49.7% on day 5 (p = 0.014) and 71.4% on day 10 (p = 0.077) after the initiation of therapy (fig. 4). Similar antitumor activity was observed in mice treated with either the unpurified virus-containing preparation or its purified formulation that included HES: on day 5, tumor growth inhibition in these groups was 31.3% (p = 0.050) and 51.5% (p = 0.065), respectively, and on day 10 it was 61.4% (p = 0.077) and 57.2% (p = 0.034).

Purified form of the virus-containing test drug stabilized with glucose proved incapable of significant inhibition of tumor growth throughout the entire observation period. The registered differences did not reach statistical significance (fig. 4; p > 0.05).

DISCUSSION

This study demonstrated a fairly high sensitivity of ACATOL model (colon adenocarcinoma in mice) to experimental treatment. The most capable preparations in the experiments were the reference cytostatic drug (cyclophosphamide) and the standard unpurified antitumor drug based on oncolytic strain of NDV, which significantly prolonged the latent period of neoplasm development and decreased the volume of the experimental tumors.

The unpurified test preparation produced an effect largely comparable to that of the reference drug, probably owing to two mechanisms: a direct oncolytic effect on tumor cells and activation of the antitumor immune response. It has previously been shown that oncolytic viruses, including various strains of NDV, selectively infect cancer cells by exploiting defects in antiviral defense and dysfunction of the interferon signaling pathway; they subsequently replicate within these cells and induce cytopathic effects [20, 21]. Moreover, virus-induced tumor cell destruction is accompanied by the release of tumorderived PAMPs/DAMPs, which promote the recruitment of innate and adaptive immune effector cells into the tumor microenvironment and enhance overall antitumor immune activity [22]. This may explain not only the reduced rate of tumor growth but also the prolongation of the latency period before tumor onset in the first experimental set.

We equalized the infectious titers of the NDV/Altai/ pigeon/770/2011 virus across all test drug formulations (table); therefore, all formulations should have produced a comparable direct cytotoxic effect. The broadly similar antitumor efficacy and immune-mediated activity of the formulations are likely attributable to the activation of antitumor immunity, provided that the viral particles retained the ability to adsorb to, enter, and replicate within tumor cells. At the same time, purified oncolytic virus preparations supplemented with HES or glucose showed limited or no efficacy, particularly at later time points..

One possible explanation for the limited or absent efficacy of the HES- or glucose-stabilized test drugs is the diminished ability to directly affect tumor cells and/or stimulate the antitumor immune response. The brief positive effect of the glucose-supplemented variant observed on days 15 and 20 of the study may be associated with initial viral replication and short-term release of oncolytically active factors. However, the degradation of this effect by day 25 indicates the duration of action of this form of the drug is insufficient.

HES and glucose were selected as stabilizers because they preserve viral infectivity during storage at different temperatures and for different periods [23]. Viral infectivity after storage was determined using Vero E6 cells. However, high glucose concentrations have been associated with reduced infectivity of influenza virus and Sendai virus in A549 non-small-cell lung cancer cells [24]. In that study, influenza A virus and Sendai virus were cultured in a high-glucose medium, and their infectious titers decreased over time. No comparable effect has been reported for HES-stabilized formulations. Therefore, the weaker antitumor activity of viral preparations supplemented with HES and glucose may be explained by the effects of these additives on the ability of the virus to infect tumor cells upon contact. The reduced antitumor activity also warrants a comparative analysis of the elimination rates of unpurified and stabilized viral preparations from the body.

In several reported experiments, viral oncolysates derived from transplanted tumor cell lines have been used as antitumor agents. Injection of viral lysates prepared from tumor cells significantly reduced the size of tumor nodules in animal cancer models [25]. In addition to virions, such oncolysates contain products of tumor-cell lysis and components of the culture medium, which may act as a protective barrier and stabilizing environment for the virus [26]. Virus purification and the use of serum-free medium have been reported to reduce the natural stability of certain baculovirus variants [26]. Hepatitis C virus has also been shown to lose infectivity in the presence of fetal bovine serum, presumably because serum interferes with viral attachment to cells; this effect is concentration-dependent [27].

These findings indicate that the efficacy of oncolytic viral preparations depends critically on the purification technology, the composition of excipients, and the potency of the viral strain. This interdependence warrants further investigation. Overall, the differences observed between the unpurified and stabilized preparation groups suggest that a structurally intact viral preparation provides a more prolonged and stable oncolytic effect, whereas simplified or modified forms may produce only transient effects or lack efficacy.

The primary objective of any virus production process is to obtain a preparation with the highest possible infectious titer. Subsequent steps — optimization of virion collection, purification, and concentration — may, however, compromise infectivity and damage viral particles, thereby reducing the efficacy of purified virus-containing preparations. Loss of infectious activity can occur at any stage. During initial clarification, the main targets are cellular debris and large aggregates, which are removed by low-speed centrifugation and membrane filtration. This step may reduce the infectious titer by up to 25% [28]. The most effective mitigation strategy is the use of 0.45-µm membrane filters. For NDV, 0.22-µm membranes are unsuitable because they can adversely affect infectivity, given that NDV virions range from 200 to 300 nm in diameter [29]. Ultracentrifugation, a more prolonged procedure that may expose viral particles to damaging shear forces, can further reduce the titer. For example, ultracentrifugation of herpes simplex virus type 6 reduces its infectivity by 90%. [28].

Thus, the weaker antitumor activity of the HES- and glucose-stabilized purified preparations may be associated with the purification procedure, particularly tangential-flow ultrafiltration through a membrane. It can be assumed that prolonged filtration increases damage to viral particles and that some virions may be retained by or lost to the membrane. In addition, HES and glucose may stabilize the virus to different extents, ultimately resulting in reduced infectivity and a diminished antitumor effect.

These observations are important for optimizing the composition and form of future experimental oncolytic drugs, as well as for determining strategies for their use in the treatment of oncological diseases.

The HES-stabilized purified test preparation was most effective during the early stages of the experiment, suggesting that it can suppress the growth of a tumor that has not yet become fully established. However, its efficacy in the later stages of the experimental tumor process was insufficient. This finding is consistent with the hypothesis that purification may have removed virion protein components required for efficient infection of tumor cells and/or for stimulation of the immune response. In particular, several studies have shown that alterations in the conformation of virion-associated surface proteins, such as the F and HN proteins, directly affect the ability of the virus to recognize and enter target cells, as well as its capacity to activate macrophages and NK cells [30]. A reduction in these functional interactions may leave the virus with limited cytotoxic activity and could, in theory, restrict its spread within tumor tissue.

CONCLUSIONS

In this study, we used a murine colon adenocarcinoma model (ACATOL; CT26 cell line). The experimental tumor was characterized by satisfactory growth dynamics and high sensitivity to cyclophosphamide, the reference antitumor drug. The observed effects included a longer latent period of tumor development following transplantation of malignant cells, inhibited tumor growth, and increased mean survival time in treated animals. The standard unpurified NDV-containing preparation showed high efficacy against this cancer model: for nearly all assessed parameters, its activity was comparable to that of cyclophosphamide. Thorough purification of the preparation, with glucose added as a carrier for the active component, resulted in a significant reduction in pharmacological activity across all assessed parameters. Modest, albeit clinically non-significant, effects were observed in animals treated with the HES-stabilized NDV-based antitumor preparation. These findings provide important data for studying the antitumor activity of virus-containing preparations and for optimizing their manufacturing technology in both biotechnological production and basic research.