Establishing the Kenya National Antivenom Quality Control Laboratory: Preclinical Efficacy Results of Four Antivenoms Against Venoms from the “Big Five” Snake Species in Kenya

TL;DR

This study evaluates the effectiveness of four antivenoms in Kenya against venom from five dangerous snake species to improve snakebite treatment.

Contribution

The study establishes a national antivenom quality control framework through preclinical testing of antivenoms in Kenya.

Findings

SAIMR polyvalent, AFRIVEN, and PANAF-PremiumTM showed strong venom-neutralizing efficacy in both in vitro and in vivo tests.

InoserpTM had poor binding and neutralizing performance, especially against Naja and Dendroaspis venoms.

The results support the need for a national antivenom quality control laboratory to ensure effective treatment options.

Abstract

Antivenom administration is currently the only therapy for snakebite envenoming. However, in sub-Saharan Africa, inadequate quality control systems have led to deficits in the availability, accessibility, efficacy and safety of regionally available antivenoms, which, in turn, hinder snakebite treatment and management in the region. To address this impediment to snakebite treatment in Kenya, this study aimed to assess the preclinical neutralising potencies of four different antivenoms previously or currently available in Kenya (SAIMR polyvalent, AFRIVEN, PANAF-PremiumTM and InoserpTM) against key snakes of medical importance in the region, towards establishing a national antivenom quality control laboratory. Venoms were extracted from the Kenyan “big five” medically important snake species: Naja ashei, Naja pallida, Naja nigricollis, Dendroaspis polylepis and Bitis arietans, and their lethal potencies were determined using a murine median lethal dose (LD50) assay. In vitro immunological assays (ELISAs and immunoblotting) and an established preclinical murine in vivo neutralisation assay (median effective dose [ED50]) were used to assess the immunoglobulin-binding and venom-neutralising efficacies of the test antivenoms. In vitro assays revealed high venom-binding titres of SAIMR polyvalent, AFRIVEN and PANAF-PremiumTM, and reactivity to a wide range of venom proteins across the different snake venoms. Contrastingly, InoserpTM antivenom had low binding titres and poor reactivity to the snake venom proteins. These findings were aligned with the in vivo results, where SAIMR polyvalent, AFRIVEN and PANAF-PremiumTM showed potent venom-neutralising efficacies against all the tested snake venoms, while InoserpTM had low potency and failed to neutralise the lethal effects of N. ashei, N. pallida and D. polylepis venoms at the manufacture-claimed doses. Based on these robust preclinical results, we conclude that SAIMR polyvalent, AFRIVEN and PANAF-PremiumTM antivenoms offer considerable potential for the treatment of envenoming by diverse medically important snakes in Kenya. The observed deficiencies with the InoserpTM product highlight the importance of (i) robust, independent preclinical antivenom efficacy testing and (ii) the value of establishing a quality control laboratory to inform local regulatory and procurement decision making.

Figures

Full text

1. Introduction

Snakebite envenoming (SBE) is a priority neglected tropical disease (NTD) that is particularly prevalent in sub-Saharan Africa (SSA), the Middle East, South and Southeast Asia, Oceania, and Latin America [1]. Globally, an estimated 5 million snakebites occur yearly, resulting in 81,000 to 138,000 deaths and nearly 400,000 permanent disabilities [2,3]. Most of these snakebite incidences are reported to occur among the world’s poorest people living in low- and middle-income countries of the tropics and subtropics [1]. In Africa, 500,000 envenoming incidences resulting in 7000 to 32,000 deaths are estimated to occur annually [4,5], with Kenya reporting around 15,000 cases, 719 deaths and 847 amputations each year [6,7]. Mass drug administration campaigns for NTDs in Kenya, where 13.1 million people (about 27.6% of the Kenyan population) participated, integrated snakebite incidence into their data collection. During the study period, 4667 cases of snakebite were identified, corresponding to an annual incidence rate of approximately 35.6 cases per 100,000 inhabitants [8]. The true burden of snakebite in Africa is likely underestimated due to reliance on health facility-based reporting, which fails to capture the majority of community-managed cases. Evidence from community surveys in Kenya and regional meta-analyses highlight underreporting of the snakebite incidence across sub-Saharan Africa [4,8]. The psychological sequelae arising from snakebite are also yet to be properly established.

Currently, the sole treatment for snakebite envenoming is animal-derived polyclonal antibodies, known as antivenom [9]. These consist predominately of either whole immunoglobulins (IgG) or Fab/F(ab′)2 fragments purified from the sera/plasma of large mammals, such as horses or sheep, hyper-immunised with (i) venom from either a single snake species (monospecific antivenom), or (ii), more commonly, multiple snake species (polyspecific antivenom) [10,11,12]. Despite the long history of the development and production of these serotherapies, the management and treatment of snakebite cases remains poor in many affected regions; largely due to the high costs, limited accessibility, variable efficacy and high incidence of adverse reactions associated with antivenom [13,14,15]. These deficits are particularly apparent in SSA, where many affected people lack access to proper care and treatment [16,17,18], with many of them seeking alternative treatment, such as traditional medicine [17,19]. Cultural beliefs, poverty, and poor access to health services also contribute to increased mortality and morbidity resulting from snakebite.

To combat snakebite morbidity and mortality, the World Health Organization (WHO) has highlighted access to safe, easily accessible, effective, and inexpensive antivenoms as a top priority to mitigate the considerable health consequences associated with SBE [20]. The WHO’s primary recommendation is that antivenoms are rigorously tested in two stages of trial before marketing for clinical use: (i) preclinical evaluation in small animal models and (ii) clinical evaluation in formal clinical trials [21]. Due to the limited clinical data on antivenom effectiveness in SSA [22] and the high cost of these trials, the preclinical testing of antivenom efficacy has historically been predominately relied upon for evaluating the quality and venom-neutralising efficacies of snakebite therapeutics. To help regulatory bodies register or license antivenoms that are effective for a country’s treatment needs, the WHO recommends that these preclinical murine protections against venom-induced lethality assays [23] be performed for every new antivenom and antivenom batch because of the variable consistency of biologic serotherapies [23].

In Kenya, the most common and severe snakebite envenoming cases are caused by several elapid snakes, most notably the black mamba (Dendroaspis polylepis), the spitting cobras (Naja pallida, Naja ashei and Naja nigricollis) and non-spitting cobras (Naja subfulva and Naja haje), and the two main viperid snakes, the puff adder (Bitis arietans) and the Kenyan carpet viper (Echis pyramidum leakeyi) [6]. Envenoming by these snakes induces either cytotoxic (oedema, blistering, tissue necrosis), neurotoxic (neuromuscular paralysis), and/or hemotoxic (bleeding, coagulopathy) effects in envenomed victims [24]. Antivenoms act by neutralising the toxic effects of snake venoms via antibody-toxin binding. However, the effectiveness of an antivenom is largely restricted to the venoms used to hyper-immunise the antivenom-producing horses or sheep [25]. Because of the extensive variation in snake venom compositions within and across snake species, based on factors such as the diet, age or sex of the snakes and their geographic location [26], the selection of venoms for antivenom production has very important implications for the effectiveness of the resulting product [27,28], particularly in the context of the polyspecific efficacy of antivenoms [29].

The antivenoms available on the Kenyan market are currently externally sourced from India and Mexico and are marketed with limited, non-independent, preclinical evidence of their efficacies. The most commonly marketed antivenoms in Kenya over the past several years have been Snake Venom Antiserum African, manufactured by VINS Bioproducts, and Inoserp^TM^ (PAN-AFRICA) by INOSAN Biopharma [30]. Both products were withdrawn from the Kenyan market in 2022 after performing poorly in our preclinical efficacy assays using Kenyan venoms [29] and failing to achieve a positive review in the WHO’s risk-benefit assessment process, and they were put under review by the Kenyan medicines regulatory body, the Pharmacy and Poisons Board (PPB), on 29 July 2022, with the body issuing safety alerts on these products [31]. VINS Bioproducts has since produced a new batch formulation (AFRIVEN) that has yet to be commercialised in Kenya. These new products by VINS and INOSAN’s Inoserp^TM^ are undergoing a risk–benefit assessment by the WHO, and, as of yet (2025), neither has achieved approval via this process [32]. Two other antivenoms of potential use on the regional market include SAIMR Polyvalent Snake Antivenom by South African Vaccine Producers, which is costly ($315 per vial) [29,30], and a new batch formulation of PANAF-Premium^TM^ produced by Premium Serums and Vaccines, which has recently been positively assessed by the WHO’s antivenom risk–benefit assessment procedure [32] and, since, approved for use in the country by the PPB. Testing the efficacies of new antivenoms or new batches being introduced into a region is of vital importance, as the consequences of importing inefficacious products inappropriate for a region can be disastrous. This was highlighted by a study conducted in Ghana that reported an increase in mortality in snakebite patients from 1.8% while using effective antivenom to 12.1% when an ineffective antivenom that had not undergone independent preclinical testing was introduced into the region [13].

Through the African Snakebite Research Group (ASRG), Kenya’s Ministry of Health established a preclinical antivenom testing facility at the Kenya Snakebite Research and Intervention Centre (K-SRIC). This collaboration enabled capacity-building activities at the K-SRIC, including: (1) the creation of a dedicated herpetarium to house venomous snakes to produce venom standards for antivenom testing, and (2) the training of staff on murine preclinical models to enable the robust evaluation of antivenom efficacy, the first of its kind in SSA. The K-SRIC, supported by the Kenya Ministry of Health, aims to be a hub of antivenom quality control in sub-Saharan Africa, adhering to the WHO-recommended guidelines to deliver informative assessments of antivenom efficacy for the region. To this end, in this study, we determined the lethal potencies of the venoms of the “big five” snakes of medical importance in Kenya (N. ashei, N. pallida, N. nigricollis, B. arietans and D. polylepis) and assessed the venom-neutralising potencies of the four antivenoms described above against each of these venoms through a series of in vitro immunological and in vivo preclinical efficacy assays. The variable efficacies of the tested products demonstrate the importance of the local testing of antivenoms and the value of establishing local quality control laboratories for evaluating treatments for snakebite.

2. Results

2.1. Antivenom Quality Control

2.1.1. Antivenom Protein Concentration

The protein concentrations (IgG content) of the tested antivenoms, measured against calibrated IgG standards, were determined in triplicate using direct UV on a Biodrop spectrophotometer, and the results are presented in Table 1. Statistical analysis using ordinary one-way ANOVA with Tukey’s multiple-comparisons test showed that all pairwise comparisons between the protein concentrations of the antivenoms were significantly different (p < 0.0001). SAIMR polyvalent had the highest protein concentration, followed by AFRIVEN. The protein concentration of PANAF-Premium^TM^ was moderate, while that of Inoserp^TM^ was substantially lower than those of the other products, representing a seven-fold decrease in protein concentration compared with SAIMR polyvalent.

2.1.2. Antivenom Purity

The purities and integrities of the antivenom immunoglobulin fragments were assessed through reducing and non-reducing SDS-PAGE conditions, with purified horse IgG used as the control (for intact immunoglobulin), as shown in Figure 1. Under non-reducing conditions, SAIMR polyvalent, AFRIVEN and PANAF-Premium^TM^ were shown to have great physicochemical purities, with their F(ab′)2 fragments presenting with molecular masses of ≥100 kDa and minimal impurities. Despite Inoserp^TM^ having minimal impurities, the immunoglobulin fragment in the vial was shown to be very faint, again reflecting the low protein concentration. The reducing conditions of the antivenom fragments showed bands of around 25 kDa, with Inoserp^TM^ bands much fainter than those detected in the other three products.

2.2. The Binding Titres of the Four Antivenoms to Key Medically Important Kenyan Snake Venoms

To provide a quantitative readout of the antivenom cross-reactivity, an end-point titration ELISA was used to determine the IgG-binding titres of the antivenoms to N. ashei, N. pallida, N. nigricollis, B. arietans and D. polylepis venoms at 405 nm absorbance (Figure 2).

N. ashei venom: SAIMR polyvalent had the highest binding titres to this venom, though there was no significant difference between SAIMR polyvalent and AFRIVEN (titre: 1:7.81 × 10^6^, p > 0.05, ODs = 0.501 and 0.395) or PANAF-Premium^TM^ (titre: 1:7.81 × 10^6^, p > 0.05, ODs = 0.501 and 0.350). Inoserp^TM^ had the lowest binding titre of 1:312,500 to N. ashei venom, with an OD value of 0.475 at that dilution.N. pallida venom: SAIMR polyvalent, AFRIVEN, and PANAF-Premium^TM^ showed no significant differences (p > 0.05) in their binding titres (1:7.81 × 10^6^) to the venom of N. pallida, with OD values of 0.402, 0.374 and 0.323, respectively. Inoserp^TM^ had the lowest binding titre of 1:312,500 (OD = 0.531) to this venom.N. nigricollis venom: There was also no significant difference in the binding titres of SAIMR polyvalent, AFRIVEN and PANAF-Premium^TM^ (titre: 1:1.56 × 10^6^, p > 0.05, ODs = 0.744, 0.746 and 0.622, respectively) to the venom of the related spitting cobra N. nigricollis. These antivenoms again showed higher binding than Inoserp^TM^, which had a binding titre of 1:62,500 (OD = 0.832 at this concentration).D. polylepis venom: All four antivenoms showed high binding capabilities to the venom of D. polylepis, with SAIMR polyvalent, AFRIVEN and PANAF-Premium^TM^ exhibiting titres of 1:7.81 × 10^6^ (p > 0.05, ODs = 0.444, 0.346 and 0.363, respectively), while INOSERP had a slightly lower titre of 1:1.56 × 10^6^ (OD = 0.324).B. arietans venom: SAIMR polyvalent, AFRIVEN and PANAF-Premium^TM^ showed high and comparable binding capabilities against B. arietans venom, with titres of 1:7.81 × 10^6^ (p > 0.05, ODs = 0.411, 0.377 and 0.346, respectively), while Inoserp^TM^ once again exhibited the lowest binding with a titre of 1:312,500 (OD = 0.424) (Figure 2).

2.3. Visualisation of Antivenom Binding to Key Medically Important Kenyan Snake Venom Proteins

Immunoblotting was used to visualise the venom protein binding by each of the antivenoms. SAIMR polyvalent and AFRIVEN antivenoms showed the greatest binding intensities to the venom proteins of N. ashei, N. pallida, N. nigricollis, B. arietans and D. polylepis, recognising a wide range of proteins, including both high- (>25 kDa) and low-molecular-weight (<25 kDa) toxins, and revealing several protein bands not detectable via SDS-PAGE (Figure 3). PANAF-Premium^TM^ had a medium binding intensity to the venom proteins of the selected medically important snakes when compared to the SAIMR polyvalent and AFRIVEN antivenoms. Inoserp^TM^ had the lowest binding intensity to the venom proteins of the five medically important snake venoms, only showing faint binding to the low-molecular-weight proteins in the venoms of N. ashei, N. pallida, N. nigricollis and D. polylepis and almost no binding to B. arietans venom proteins. Purified horse IgG, which was used as a negative control, showed no specificity to the venom proteins of the key snakes of medical importance. Overall, these results likely reflect the consequences of the substantial differences in the protein contents of the four tested antivenoms.

2.4. Venom Potency Profiles of Medically Important Kenyan Snakes

The intravenous lethal potencies of the five snake venoms were assessed using the murine LD_50_ assay and were found to be 8.77 µg/mouse for N. pallida; 11.13 µg/mouse for N. nigricollis; 14.60 µg/mouse for N. ashei; 5.36 µg/mouse for D. polylepis; and 15.65 µg/mouse for B. arietans, as shown in Table 2. The lethal potencies were compared using overlapping confidence intervals. This study revealed a significant difference in the potencies of the key medically important snake venoms, with D. polylepis showing the highest potency, three times more lethal than B. arietans, which showed lower potency than all four elapid venoms in this model.

2.5. In Vivo Venom-Neutralising Potencies of SAIMR Polyvalent, AFRIVEN, PANAF-PremiumTM and InoserpTM Antivenoms

The neutralising potencies of the four antivenoms against the lethal effects of the key medically important snakes in Kenya were assessed using the ED_50_ assay. These potency results were expressed as LD_50_/mL (number of LD_50_ completely neutralised per unit volume of antivenom) and are shown in Figure 4, with other ways of reporting the ED_50_ data found in the Supplementary Materials data.

Neutralisation of N. nigricollis venom: All four antivenoms were able to neutralise the lethal effects of N. nigricollis venom as per the manufacturer’s claims, ≥20 LD_50_/mL (AFRIVEN and PANAF-Premium^TM^) and ≥50 LD_50_/mL (Inoserp^TM^). South African Vaccine Producers makes no market claim regarding the potency of its product SAIMR Polyvalent. However, the new batch formulations of AFRIVEN and PANAF-Premium^TM^ showed greater neutralising potencies (140.16 and 112.31 LD_50_/mL, respectively) than SAIMR polyvalent (54.27 LD_50_/mL) and Inoserp^TM^, which failed to neutralise the 5× LD_50_ of N. nigricollis but gave a neutralising potency of 62.62 LD_50_/mL with a reduced 2.5× LD_50_ venom challenge dose.Neutralisation of N. ashei venom: PANAF-Premium failed to neutralise the 5× LD_50_ dose of N. ashei but gave a neutralising potency of 93.36 LD_50_/mL with a reduced 3× LD_50_ challenge dose. AFRIVEN and SAIMR Polyvalent were equally effective at neutralising the lethal venom effects of N. ashei (56.64 and 56.99 LD_50_/mL, respectively). Inoserp^TM^ had the lowest neutralising potency of 22.60 LD_50_/mL against N. ashei with a reduced 2.5× LD_50_ dose, as it failed to neutralise its lethal venom effects at 5× LD_50_.Neutralisation of N. pallida venom: SAIMR polyvalent and AFRIVEN were equally effective at neutralising the lethal effects of N. pallida (94.64 and 88.71 LD_50_/mL, respectively), followed by PANAF-Premium^TM^ (52.57 LD_50_/mL). Inoserp^TM^ antivenom was less potent and only neutralised the lethal effects of N. pallida with a potency of 21.44 LD_50_/mL with a reduced 2.5× LD_50_ dose. It failed to neutralise the 5× LD_50_ dose of N. pallida at the highest permissible antivenom volume of 100 µL.Neutralisation of B. arietans venom: Inoserp^TM^ also failed to neutralise the 5× LD_50_ of B. arietans but gave a potency of 89.01 LD_50_/mL with a reduced 2.5× LD_50_ dose. SAIMR polyvalent antivenom performed better than AFRIVEN, PANAF- Premium^TM^, and Inoserp^TM^ in neutralising the lethal effects of B. arietans, with the highest neutralising potency of 1600 LD_50_/mL. However, AFRIVEN (516.81 LD_50_/mL), PANAF-Premium^TM^ (191.31 LD_50_/mL), and Inoserp^TM^ (89.01 LD_50_/mL) were all capable of neutralising the lethal effects of B. arietans venom, and while there was considerable variation in their potencies, all met the manufacturer’s neutralisation claims.Neutralisation of D. polylepis venom: SAIMR polyvalent neutralised the lethal effects of D. polylepis at a higher potency (535.45 LD_50_/mL) than AFRIVEN (200.00 LD_50_/mL) and PANAF-Premium (102.99 LD_50_/mL), but all three met the marketed potency threshold claims. Contrastingly, Inoserp^TM^ failed to neutralise the 5× LD_50_ dose of D. polylepis and only neutralised its lethal venom effects with a low potency value of 34.14 LD_50_/mL at a reduced 2.5× LD_50_, which did not meet the marketed claims.

Overall, SAIMR polyvalent, AFRIVEN, and PANAF-Premium^TM^ antivenoms were able to neutralise the lethal effects of the five medically important snakes. Inoserp^TM^ antivenom failed to neutralise the lethal effects of the selected snake venoms at the 5× challenge dose. The challenge dose was reduced to 2.5× LD_50_, where Inoserp^TM^ was effective against B. arietans venom and as effective as SAIMR polyvalent against N. nigricollis venom but failed to neutralise the lethal effects of N. ashei, N. pallida and D. polylepis.

3. Discussion

Antivenom therapy is currently the most effective treatment for snakebite envenoming [9], but there are crises in both the antivenom availability and efficacy in sub-Saharan Africa [15]. Most antivenoms used in this region are sourced from countries outside the African continent, such as India, Mexico, Spain, and Costa Rica [22]. South Africa is the only African country manufacturing snake antivenom, with its products being marketed mainly in the southern countries of the continent [22]. Most of the commercial antivenoms in SSA are marketed with limited data on their clinical safety or efficacy and thus lack proper quality control, and some of these products have proven to be dangerously ineffective against regional snakes, which leads to low confidence in the imported products [14,33,34]. For quality control purposes, the WHO recommends the preclinical assessment of new antivenom batches. Despite this, new batches with minimal available preclinical efficacy data are being introduced into the region [35]. As part of the African Snakebite Research Group (ASRG) remit, the K-SRIC in collaboration with the Liverpool School of Tropical Medicine (LSTM) [29] and the LSTM in collaboration with the Eswatini Research and Intervention Centre (E-SRIC) [36] previously performed independent antivenom quality control studies for the African continent, and the outcomes from these studies are used to help inform regulatory and procurement decisions related to antivenom suitability. This study continues this research objective.

Our results of the venom LD_50_ potencies of Kenyan snakes maintained in the K-SRIC’s herpetarium showed notable differences to those reported for the same species sourced from other countries. Kenyan N. nigricollis venom was found to be two times more potent than the previously reported LD_50_ of N. nigricollis (24.40 μg/mouse) from Tanzania [29]. The LD_50_ of our Kenyan N. pallida venom (8.77 µg/mouse) was in line with our previous findings [29], where the LD_50_ was found to be 9.29 µg/mouse. However, a study by Petras and colleagues [37] found Kenyan N. pallida venom to be half as potent (LD_50_ of 17 µg/mouse), suggesting, perhaps, that intraspecific variation may exist. The Kenya D. polylepis LD_50_ determined here was comparable with prior reports from Tanzanian black mambas [29]. Similarly comparable findings were obtained with prior reports of Kenyan B. arietans venom [29], though venom from Kenyan puff adders was found to be two times more potent than B. arietans venom from Eswatini, which had an LD_50_ of 32.5 µg/mouse [36]. Venom from Kenyan N. ashei was more potent than venom from its close relative, N. mossambica, which was previously reported to have an LD_50_ of 22 μg/mouse [37]. These findings of intraspecific variation in venom potency highlight the importance of local-specific testing for determining the neutralising capacities of antivenom vials and choosing appropriate batches for use in the treatment of snakebite envenoming.

We used a series of in vitro and in vivo tests to evaluate the ability of four antivenoms to neutralise the venoms of medically important Kenyan snakes. These tests included ELISAs and immunoblotting to measure the ability of each antivenom to bind to specific venom components, followed by the ED_50_ assay to assess their ability to neutralise the lethal effects of snake venom in a murine model. ELISA results demonstrated that SAIMR polyvalent, AFRIVEN and PANAF-Premium^TM^ antivenoms showed high IgG-binding titres to all the Kenyan venoms (N. ashei, N. pallida, N. nigricollis, B. arietans and D. polylepis). This is in line with the findings of a recent study [29] that demonstrated SAIMR Polyvalent’s high IgG titres to the venom of D. polylepis and B. arietans. The Inoserp^TM^ antivenom batch used in our study had the lowest IgG-binding titres to all the Kenyan snake venoms, which may be the result of the low antibody content. In the immunoblotting experiments, SAIMR Polyvalent and AFRIVEN showed high binding intensities to the full diversity of proteins in all five Kenyan snakes. PANAF-Premium^TM^ also recognised diverse venom proteins found across the different venoms, but with a more moderate binding intensity and perhaps recognising fewer of the less abundant venom proteins. Correlating with the ELISA results, Inoserp^TM^ had the lowest binding intensities, especially to B. arietans venom proteins, and only bound to the low-molecular-weight proteins present in N. ashei, N. pallida, N. nigricollis and D. polylepis venoms.

In the in vivo neutralisation assays, the neutralising potencies of the antivenoms were assessed based on the manufacturers’ claims (the neutralising potency of SAIMR Polyvalent is not reported by the manufacturer; thus, results were compared to those of the other antivenoms). As per the manufacturer, 1 mL of AFRIVEN and PANAF-Premium^TM^ is said to neutralise not less than 20 LD_50_ of N. nigricollis venom, 25 LD_50_ of D. polylepis and B. arietans, with the former also stating paraspecificity to N. pallida and N. ashei venoms. The marketed potency of Inoserp^TM^ is said to be at least 50 LD_50_ per mL against N. nigricollis, D. polylepis and B. arietans venoms. SAIMR polyvalent antivenom showed the highest neutralising potencies against the lethal effects of B. arietans and D. polylepis venoms, requiring low volumes of antivenom to prevent venom-induced lethality. These findings are in line with those of recent studies [29,36] that tested the preclinical efficacies of African commercial antivenoms and showed the high neutralising potency of SAIMR polyvalent antivenom against B. arietans and D. polylepis venoms. AFRIVEN and PANAF-Premium^TM^ were also effective at potently neutralising the venoms of B. arietans and D. polylepis but required higher antivenom volumes than SAIMR polyvalent. In terms of the spitting cobras, SAIMR polyvalent and AFRIVEN showed equal neutralising potencies against the lethal effects of N. ashei and N. pallida, while AFRIVEN and PANAF-Premium^TM^ were the most effective at neutralising the lethal effects of N. nigricollis venom, with greater neutralising potencies than that of SAIMR Polyvalent. These findings could be attributed to the fact that N. nigricollis venom was used in the immunisation process for the production of AFRIVEN and PANAF-Premium^TM^ antivenoms but not for SAIMR polyvalent (which uses N. mossambica for immunisation instead).

One notable finding was that the new formulation of the VINS antivenom (AFRIVEN) showed greater neutralisation potency against the venoms of key snakes of medical importance compared to the previous formulation, which, in 2017, was shown to have low neutralising potency and lacked specificity to regionally relevant snakes [29]. This corresponds with an increase in the total protein concentration per vial of this newly formulated antivenom, which could explain the improved in vivo efficacy observed. While the PANAF-Premium^TM^ antivenom showed the highest neutralising potencies against N. ashei venom, these experiments were performed at a 3× LD_50_ challenge dose, due to a lack of protection in the 5× LD_50_ venom challenge. Thus, despite the lower reported potencies, the SAIMR polyvalent and AFRIVEN antivenoms were the most effective against this species, as both gave comparable levels of effective neutralisation against the 5× LD_50_ venom challenge dose. Highlighting the relative lack of efficacy, the Inoserp^TM^ product was ineffective at neutralising the lethal effects of N. ashei, N. pallida, N. nigricollis, B. arietans and D. polylepis venoms at 5× LD_50_ challenge doses, requiring us to use reduced challenge doses of 2.5× LD_50_. Even with this reduced venom challenge, Inoserp^TM^ antivenom showed the lowest neutralising potencies against the tested snake venoms, with the exception of equivalence with SAIMR polyvalent in neutralising the lethal effects of N. nigricollis, though the SAIMR polyvalent experiments were conducted with 5× LD_50_ venom challenge doses. Overall, our in vivo findings align with the described in vitro results, where Inoserp^TM^ showed poor venom recognition capabilities towards all the snake venom proteins and low IgG-binding titres, and, as mentioned, the low efficacy of Inoserp^TM^ seems likely to be attributed to its low total IgG content per vial, in line with our previous findings [29]. It is worth noting that the batch of SAIMR polyvalent antivenom used in this study had been expired for 3 years prior to use but still showed promising neutralising potencies against the key snakes of medical importance in Kenya. In line with these findings, a recent study performed by Solano and colleagues [38] demonstrated the retained potency of expired SAIMR polyvalent antivenoms, even after more than 10 years post-expiry.

While the above preclinical results meet the WHO guidelines and serve to provide local national regulatory agencies with the information they need, there is still a strong need for these in vivo murine antivenom assessments to be complemented by additional research activities. We recommend further studies to profile snake venom inhibition using in vitro bioactivity assays [e.g., phospholipase A2, snake venom metalloproteinase, blood coagulation, fibrinogenolytic, nicotinic acetylcholine receptor antagonism, cell cytotoxicity assays, among others [28,36]] and additional preclinical tests to assess the antivenom neutralisation of specific venom-induced pathologies associated with morbidity, such as dermonecrosis (Minimum Necrotic Dose [MND]) and haemorrhage (Minimum Haemorrhagic Dose [MHD]) [23]. Of critical importance is performing downstream clinical observational studies or, in an ideal world, formal clinical trials, designed to robustly evaluate antivenom efficacy in real-world snakebite patients. In Cameroon, clinical studies have been conducted to assess the effectiveness of antivenoms used in some health centres in the northern and southern parts of the country. In one study, Inoserp PAN-AFRICA was shown to be clinically effective in the management of snakebite envenoming, which was predominantly caused by viperid species (Echis spp.), with rapid improvement in coagulation disorders and neurotoxicity following antivenom administration [39]. Similarly, in another study, PANAF-Premium^TM^ was found be safe and clinically effective under routine treatment conditions, with most treated bites being from Echis species [40]. Such studies demonstrate the context-dependent effectiveness of African polyvalent antivenoms, likely reflecting geographical differences and species-specific venom profiles. The K-SRIC, in collaboration with the LSTM under the African Snakebite Research Group (ASRG), has also been carrying out clinical research studies focused on defining the pathophysiology of envenoming to help guide future therapy and diagnosis evaluations [41]. Finally, there remains a need to develop antivenoms locally to address the issues associated with the availability, cost, regional specificity and efficacy of currently available antivenoms. The Kenya Ministry of Health through the K-SRIC is in the process of creating snake venom standards and leading the development of the first local antivenom for use in the East African region.

4. Conclusions

The findings of this study demonstrate the preclinical neutralising potencies of PANAF-Premium^TM^, AFRIVEN and SAIMR Polyvalent antivenoms against the venoms of several medically important snakes in Kenya, namely, N. ashei, N. pallida, N. nigricollis. B. arietans and D. polylepis. Inoserp^TM^ showed the lowest potency both in vitro and in vivo at neutralising the lethal effects of these snake venoms. While SAIMR polyvalent often showed the highest venom-neutralising potencies, the high cost and poor availability of this product [29] has and still does restrict its potential use in East Africa. Promisingly though, the two new antivenom formulations evaluated here (AFRIVEN and PANAF-Premium^TM^) showed improvement from the previous versions of the antivenoms produced by the same manufacturers, with generally high neutralising potencies across the tested venoms.

Our findings highlight variations in antivenom efficacy, underscoring the need for rigorous and continuous quality control tests for new antivenom formulations and batches. Venom reference standards are a critical component of antivenom quality control, as they provide a benchmark for the comparison of results across manufacturers and studies and facilitate regulatory evaluation. The results of this study will contribute to the data-informed selection of antivenoms for national treatment needs by ensuring that only effective antivenoms robustly evaluated via independent preclinical testing are stocked in healthcare facilities. Having readily available and effective antivenoms will make a major contribution to decreasing snakebite-associated morbidities and mortalities and thereby contribute to improving public health associated with the world’s most lethal neglected tropical disease.

5. Materials and Methods

5.1. Ethical Statement

The animal study protocols for carrying out the preclinical lethal dose 50 (median lethal dose, LD_50_) assay, effective dose 50 (median effective dose, ED_50_) assay, snake restraint and venom sample collection were approved by the Institutional Scientific and Ethics Review Committee of the Kenya Institute of Primate Research (KIPRE/ISERC) and granted under approval ISERC/04/2017. Other necessary research permits and licenses were also sought. These include Prior Informed Consent and Mutually Agreed Terms under the Nagoya Protocol on Access and Benefit Sharing, Wildlife Research and Training Institute/Kenya Wildlife Service research permit number—WRTI-0264-01-23—and the National Commission for Science, Technology and Innovation research license number—NACOSTI/P/19/1012. The WHO guidelines for the Production, Control and Regulation of Snake Antivenom Immunoglobulins [23] were used to guide experiments performed to determine the LD_50_ of snake venoms and ED_50_ of antivenoms. All aspects of the study design, experimental animals, blinding, critical reagent details, experimental procedures, statistical analysis and results were aligned with the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines 2.0 to ensure proper reporting, reproducibility, and transparency and proper animal research standards [42].

5.2. Antivenoms

The AFRIVEN antivenom used in this study is a new batch formulation yet to be marketed in Kenya and was donated by the manufacturer (VINS Bioproducts Ltd., Hyderabad, Telangana, India) for evaluation. PANAF-Premium^TM^ is also a new batch formulation already commercialised in Kenya and was donated by the manufacturer (Premium Serums and Vaccines PVT. Ltd., Narayangaon, Maharashtra, India). The Inoserp^TM^ antivenom was a batch withdrawn from the Kenyan market after our preclinical studies had commenced and was purchased independently. The SAIMR polyvalent snake antivenom (South African Vaccine Producers (SAVP) PTY., Johannesburg, South Africa) was donated by the Liverpool School of Tropical Medicine. More details on the antivenoms used are shown in Table 3. All antivenoms are manufactured as F(ab’)2 fragments of equine-derived IgG.

5.3. Snake Collection and Venom Preparation

With the approval of the Kenya Wildlife Services (KWS), all the snakes used in this study (large brown spitting cobras (N. ashei), red spitting cobras (N. pallida), black-necked spitting cobras (N. nigricollis), black mambas (D. polylepis) and puff adders (B. arietans)) were captured from the wild (Figure 5) using snake-handling tools, given identification codes and brought to and maintained in the well-ventilated, biosecure K-SRIC herpetarium in Nairobi. Upon arrival, the snakes were kept in individual, appropriately labelled cages and quarantined for a minimum of six weeks. Daily health checks were performed, and the cages were regularly cleaned and provided with appropriate environmental enrichment. The snakes were provided with water and fed within the appropriate time (usually once a week). After quarantine, the snakes were individually caged in well-ventilated, secure snake rooms appropriate for the species.

For venom extraction, the snakes were removed from the cages with appropriate hooks, secured on the floor (N. ashei, N. pallida, N. nigricollis and D. polylepis) or bench (B. arietans) and gripped securely by hand. The snake was then allowed to bite and inject venom into a parafilm-covered, labelled, sterilised beaker and was assisted by massaging the venom glands using well-established K-SRIC protocols. The collected venoms were immediately frozen (−20 °C) and then lyophilised using a freeze dryer (Scanvac Coolsafe Freeze Dryer, LaboGene, Allerød, Denmark) and stored as a powder at 4 °C. Crude venoms were obtained from adult specimens of N. ashei (n = 5), N. pallida (n = 4), N. nigricollis (n = 4), D. polylepis (n = 4) and B. arietans (n = 5) to create the venom batches for this study.

5.4. Antivenom Protein Concentration

To determine the total IgG content of each antivenom vial, antivenoms were first reconstituted as per the manufacturer guidelines (for lyophilised antivenoms) and then diluted to 1:250 in PBS and measured in triplicate using the A280 direct-UV method (BioDrop µLITE, Biochrom Ltd., Cambridge, UK). To ensure equipment accuracy, purified horse IgG (Bio-Rad, Hercules, CA, USA) standards of known values (1000, 750, 500 and 250 μg/mL) were used as standards.

5.5. Antivenom Purity (SDS-PAGE Profiles)

A previously described method [29], with minor modifications, was used to evaluate the antivenom purity. Ten microlitres of a 1:200 dilution in PBS of each antivenom was reconstituted in 2X protein-loading buffer (1.25 M Tris pH 6.8, 4% SDS, 20% glycerol, 2 mg bromophenol blue) with β-mercaptoethanol (reduced) and without β-mercaptoethanol (non-reduced), heated at 85 °C for 5 min and separated in 15% separating gel at 200 volts for 1 h 30 min. A PageRuler™ Plus prestained protein ladder 10 to 250 kDa (ThermoFisher Scientific, Vilnius, Lithuania) was used as a standard marker. The resultant proteins were stained with Coomassie Blue R-250 and visualised using a gel documentation reader (Uvidoc HD6, UVITEC Ltd., Cambridge, UK).

5.6. In Vitro Immunological Assays

5.6.1. SDS-PAGE and Immunoblotting

For SDS-PAGE gel electrophoretic venom separation and antivenom immunoblotting, we used previously described methods [29,36] with minor modifications. Whole venom (10 μg of 1 mg/mL in PBS) was reconstituted in 2X protein-loading buffer with β-mercaptoethanol, heated at 85 °C for 5 min and separated in a 15% separating gel at 200 volts for 1 h 30 min. A PageRuler™ Plus prestained protein ladder 10 to 250 kDa (ThermoFisher Scientific, Vilnius, Lithuania) was used as a standard marker. The resultant proteins in the gels were transferred onto 0.45 μm nitrocellulose membranes (GE Healthcare, Germany) and run for 3 h at 70 volts. Following successful protein transfer, the membranes were blocked in 5% non-fat milk in Tris-buffered saline with Tween 20 (TBS-T) (0.01 M Tris-HCL pH 8.5; 0.15 M NaCl; 1% Tween 20) for 45 min at room temperature on an orbital shaker. The membranes were washed 3 times for 5 min in TBS-T on an orbital shaker and were then incubated with antivenoms or purified horse IgG (Bio-Rad, Hercules, CA, USA) at a 1:5000 dilution in blocking solution overnight at 4 °C. After 16 h, immunoblots were washed as above and incubated with horseradish peroxidase-conjugated rabbit anti-horse IgG secondary antibody (Sigma-Aldrich, Saint Louis, MO, USA) at a 1:2000 dilution in blocking solution for 2 h at room temperature. The immunoblots were washed with TBS-T as above, and the binding of antivenom proteins to the venom was visualised by the addition of DAB substrate (50 mg 3,3-diaminobenzidine, 100 mL PBS and 0.024% hydrogen peroxide) and photographed using a gel documentation reader (Uvidoc HD6, UVITEC Ltd., Cambridge, UK).

5.6.2. End-Point Titration Enzyme-Linked Immunosorbent Assay

For the end-point ELISA experiments measuring venom–antivenom binding, we followed previously described methods [43] with minor modifications. One hundred nanograms of selected snake venoms was prepared in carbonate–bicarbonate coating buffer (pH 9.6, Sigma-Aldrich, Saint Louis, MO, USA) and 100 µL was added to each well of ELISA 96-well plates (Nunc, Denmark) and incubated overnight at 4 °C. After 16 h, the plates were washed six times with TBS-T (0.01 M Tris-HCL pH 8.5; 0.15 M NaCl; 1% Tween 20) and then incubated with 5% blocking solution (non-fat milk in TBS-T) for 3 h. After blocking, the plates were washed as above. The antivenoms (SAIMR polyvalent, AFRIVEN, PANAF-Premium^TM^, and Inoserp^TM^) were added to the plates in duplicate at an initial dilution of 1:100 in blocking solution and serial-diluted at fivefold across the plate. Purified horse IgG (Bio-Rad, Hercules, CA, USA) was used as a negative control. The plates were incubated overnight at 4 °C with the primary antibodies. After 16 h, the plates were washed six times again with TBS-T and incubated with horseradish peroxidase-conjugated rabbit anti-horse IgG secondary antibody (1:1000; Sigma-Aldrich, Saint Louis, MO, USA) in PBS for 2 h. After the plates were washed a final time, substrate (0.2% 2,2/-azino-bis (2-ethylbenzthiazoline-6-sulphonic acid) in 0.05 M citric acid buffer, pH 4.0, containing 0.015% hydrogen peroxide (Merck, Germany) was added for 25 min, and the optimal density was measured at 405 nm using a LT-4500 plate reader (Labtech International Ltd., Uckfield, UK). The antivenom titres were defined as the dilution at which the mean absorbance values were above the negative-control mean absorbance (1:100 dilutions) plus two times the standard deviation [25,43]. ELISA data was analysed using GraphPad Prism software version 8.4.3 (686). Statistical comparisons of antivenom titres were performed using two-way analysis of variance (ANOVA) with Tukey’s multiple-comparisons test. p values < 0.05 were considered to have statistically significant differences.

5.7. Mice

White male CD-1 mice weighing 18–22 g were used in this study, sourced from breeding pairs purchased from Charles River Laboratories (Calco, Italy) that were then maintained and bred at the K-SRIC rodent facility. The animals were housed in biosecure rooms with appropriate conditions of 12 h light/darkness, ventilation rates of 10 to 15 air changes per hour, temperature ranges of 17.8 to 26.1 °C, humidity levels between 45 and 60% and noise levels below 70 decibels (dB). The mice had ad lib access to clean water and mouse food pellets (Unga Feeds Ltd., Nakuru, Kenya). Their enrichment consisted of soft, sterile wood shavings, pipes, tissue and ground nuts.

5.8. Preclinical Murine Assays

Animal Models of Venom-Induced Mortality and Antivenom Efficacy

The WHO guidelines [23] for the preclinical evaluation of venom lethality and antivenom efficacy were used to guide the following assay designs.

5.8.1. In Vivo Lethality Assay (Median Lethal Dose 50, LD50)

The median lethal dose (LD_50_) (the amount of venom that caused lethality in 50% of injected mice) assay was used to determine the lethal potency of each of the snake venoms. Mice were placed into groups of 6 and 100 µL of varying venom doses prepared in 1X phosphate-buffered saline (1X PBS, pH 7.2) was intravenously injected into the tail veins. As required by the model design, one control group of six mice was intravenously injected with 100 μl of 1X PBS. The death and survival of the mice in each group were recorded over 24 h. The venom LD_50_ and 95% confidence limits were calculated by probit analysis.

5.8.2. In Vivo Neutralisation Assay (Median Effective Dose 50, ED50)

The median effective dose (ED_50_) (the amount of antivenom that enabled survival in 50% of mice injected with antivenom–venom mixtures) assay was used to determine the venom-neutralising potencies of the antivenoms. Varying doses of each antivenom were reconstituted with the 5× venom challenge dose (5× LD_50_), brought to a total volume of 200 μL with 1X PBS, and incubated at 37 °C for 30 min before injection. The antivenom–venom mixture was injected intravenously into the tail veins of the mice (group size, n = 6). The venom challenge doses were reduced to 2.5× LD_50_ for Inoserp^TM^ against all test venoms, and to 3× LD_50_ for PANAF-Premium^TM^ against N. ashei venom, as these antivenoms were ineffective at the highest permissible antivenom volume (100 μL) when using the 5× LD_50_ challenge dose. The mortality and survival of the mice in each experimental group were recorded over 24 h. The antivenom ED_50_ and 95% confidence were calculated by probit analysis. The resulting antivenom neutralising potency was calculated using the following formula [43,44,45]:

$$[eqn]$$

CD represents challenge dose

$$[eqn]$$

5.9. Data Analysis

ELISA data was statistically analysed using two-way ANOVA in GraphPad Prism software version 8.4.3 (686). Probit analysis using IBM SPSS Statistics version 20 was used to generate the venom lethal dose (LD_50_), median effective dose (ED_50_) and corresponding 95% confidence interval limits. Interval estimates for the dose limits (LD_50_ and ED_50_) were generated based on the 95% CI. The overlap of the 95% confidence intervals was used to determine the significance of the LD_50_ value variations, with values that correspond to non-overlapping confidence intervals being viewed as significantly different.