Background:

This is the first study to examine a cohort that engages in the practice of immunization with snake venoms. In this practice, either fresh wet venom or venom reconstituted from freeze-dried form is used in vaccination protocols to produce hyper-immunity to venom.

Methods:

This is a retrospective community-initiated collaborative research (CICR) project that collated the records of venom immunization. Records of schedules, formulations, photographs, medical records, and diaries were collated from existing practitioners and evaluated by inspection and interviews. One accidental bite was observed over 3 days, with vital signs, and photographic records of swelling taken to verify reality of the bite. Over 74 snake-genera man-years, and 24 man-years of injection data from 8 participants, for 22 species of venomous snakes from Elapidae and Viperidae are represented. Six of those participants had detailed records of date, dose and effects.

Results:

IgG titers to 6 venoms for 4 cohort members tested of 8 included 2 with clear hyper-immune status. IgE titers were elevated for some. In 861 injections, records showed a rate of atopy/anaphylaxis of 4.3%, an infection rate of 0.58% and an abscess rate of 1.51% . Serious adverse reactions were rare and these appeared to be linked to overly aggressive immunization schedules and formulation accidents. We note that greater cross-immunity of IgE over IgG is suggested. Two basic protocols were followed, one was an approximate one month interval, the other was one or more injection(s) per week. In 176 envenomations, 175 were without antivenom treatment, two hospitalizations occurred, and one received full antivenom treatment. Dry bites were not included in our dataset. Envenomations showed a 1.14% rate of atopy/anaphylaxis, a 0.57% rate of infection and a 1.7% rate of abscess.

Conclusions:

Immunization of humans to snakebite is effective, and reasonably safe with care. Injection records suggest immune cross-reactivity between ophidians within the same family, and better cross-reactivity within the same genera. A cohort participant was pronounced dead based on EEG, and then recovered without treatment. A neurotoxin case with “brain death” EEG should stay on life support for 6 weeks to allow time for the immune system to clear venom.

Keywords: Venom immunization, Self-immunization, Envenomation, Bill Haast

1.1. Historical literature of venom immunization

There is a long history of humans using venoms to vaccinate against bites of venomous snakes. This recorded history of immunization against snake venoms goes back to Roman historians. In modern times, scientific research stopped in 1965 with Canan and Flowers while amateurs kept the practice going (Table 1). In this study, we present the first cohort of people vaccinating to snake venoms.

The colloquial term for this practice is self-immunization or SI. Herein, we will use the more precise terms of venom immunization or vaccination.

1.2. Community-Initiated Collaborative Research (CICR) venom cohort

The cohort was named Venom Immunization Protocol Research on Basic Immunology Therapeutic Experimental Method (VIPRBITEM). There are parallels and differences between this study and community-based participatory research (CBPR) [14]. The primary difference of CICR from CBPR is that members of the cohort initiated the study, defined its protocols, decided independently to perform the procedures, and provided the data.

A consent form was created and signed by members of the cohort. Consent was based on the Harvard open IRB consent and guidelines [15]. IRB approval was obtained for the collection of blood samples for laboratory analysis and for collection of data, from the Institute for Regenerative and Cellular Medicine (IRCM) Santa Monica, California (IRB approval number: IRCM-2019-209) (See supplement for expanded discussion).

1.3. Reasons for immunization with venom

We asked members of the cohort why they wanted to go through the process of becoming immune to venom. The reasons they gave are presented inTable 2.

2.1. Hyper-immunity vaccination series protocols

The VIPRBITEM cohort uses protocols for performing venom immunization based on experience handed down orally over roughly 70 years, with input from a few papers [22]. The principles are basically the same as insect venom desensitization [23]. No members of this cohort use filtration to remove bacteria nor have any attempted a toxoid, although both are discussed in Weiner’s paper [8], which is known to the cohort.

Injections are done either intramuscularly or subcutaneously in the VIPRBITEM cohort. Weiner recorded 2 intradermal venom injections. Our recommendation is to use the subcutaneous route because it exposes muscles to less damage if a dose is miscalculated. Intradermal injections are excellent for immune system presentation but may leave a scar and maximize pain.

The most common injection locations are in the thigh and abdomen above the belt line. The forearm and upper arm are less often used. Areas near joints should be avoided, as should hands, feet, and fingers.

Antihistamines are used on an as-needed basis, and there is no reason not to take them pre- or post-injection. Most practitioners have an epinephrine injection kit available and are encouraged to use it, but so far no venom vaccinator we are aware of, in or out of this cohort, has used one when warranted. Most discuss what they do with their physician and get kidney and liver blood work done from time to time. The cohort was informed that the use of either antihistamine or epinephrine will not compromise immune response, which is a common question.

We recommend use of the curated dataset created for the companion meta-analysis article [24] to this one, and the worked examples inTable 1 of this article’s supplement.

The hyper-immunization protocol consists of two phases, i.e., immunoglobulin (Ig)G generation and maintenance. There are two basic types of protocol: long-interval and short-interval ones.

2.2. Phase II: Booster maintenance

Booster dose is chosen based on feel but typically does not exceed 60% of the lethal dose 50% (LD50) for the snake venom, although some are more than LD50. Some practitioners set their boosters at fairly low doses, on the order of one milligram. Some just continue their dilute inoculations (even on a weekly basis) at much lower doses. Others inject amounts that are quite significant, as seen in the graphs that follow. Low dose boosters have been injected as often as two or more times per week for life. High dose boosters are typically injected on an approximate 3-month basis. This low-dose frequent booster schedule is not optimal immunologically but it may have other interesting effects.

3.1. Cohort composition and background

From the VIPRBITEM cohort of 10 people, there are data from a total of 9, all male, in our dataset (Figure 1), V0001-9. However, V0002 and V0009 did not have complete datasheets. The youngest age when immunizations started was 23, and the oldest member of the cohort was 52. The mean average age is 35. Defining an age for the experiment is complicated because it varies from a rough 20-year period to 7 years.

All participants in this cohort are self-selected participants who came from an online forum. Participation in the cohort was decided based on the primary investigator having discussions with participants, examining records, and judging the participants to be forthcoming on all aspects. Seven of the participants were located in North America, one was in Europe, one in Japan, and one in Southern Africa. Ethnicity was not recorded. All were self-reported in good health with no significant medical history that was not mentioned in the supplement. This is a CICR study, entirely voluntary, initiated by the cohort community.

The only female participant withdrew after being advised of teratogenic risk. One male participant died of unrelated causes during the study. The study used records kept by the participants over a period of two decades. Blood samples were only collected from participants located in the USA who volunteered to do so.

3.2. Data collection

Hyperimmunity data compiled by the VIPRBITEM cohort covering two decades were collected in our collaboration over a period of 8 years. Readers will see that these data have the ring of authenticity. Results reported contain the irregularity and idiopathic exceptions that tend to characterize real field data. In certain instances, photographic evidence and medical records were supplied on request.

The primary data collected were dose, date, pain, and swelling. In addition, notes on anaphylaxis, infection, and other health matters were collected. These immunizations represent over 68 snake-genera man-years and over 22 simple man-years of experience.

3.3. Snake venoms and dose toxicity in this hyper-immunity cohort

Venom values for minimum lethal dose (MLD) and LD50 wet and dry values are taken from the supplementary venom dataset provided using the algorithm of Hanley and Gross [24] to determine MLD within a set of studies, LD50, LD range, and LD 50% midpoint. Detailed discussion is provided in Supplement Section 1.

Elapids and viperids comprise all of the venoms in our dataset.

3.4. ELISA testing of cohort sera

ELISA plates (Millipore-Sigma) were made forC. scutulatus,C. atrox,Bothrops atrox,Naja annulata,N. kaouthia, andD. polylepis venoms (MToxins Venom Lab LLC, Oshkosh, WI; Reptile Rescue, Ranger, TX). Four cohort sera were tested on each venom plate together with normal control serum and Clodomiro Picado antivenom (AV) (Polivalente Anti-Botropico Anti-Crotalico Anti-Laquesico, Fundacion UCR Universidad Costa Rica). For IgG detection, Goat Anti-Horse IgG (ab102396, Abcam) was used for the AV wells and Goat Anti-Human IgG (AP112P, Sigma-Aldrich) for human sera. For IgE detection, Goat Anti-Human IgE (GTX77496, GeneTex) was used. Serial 4X dilutions of sera started at 100 with final dilution of 1.6384E6. AV dilution ended at 4.096E5. Plates were read on a microplate reader (MR9600T, Accuris) at 450 nm after HRP development. Signal was analyzed using Excel and a logistic regression (Equation 1). Titer positive/negative threshold (PNT) was selected conservatively based on the end of linear range. Intersection between fitted curve and PNT was calculated using equation 2.

Equations I and II: a = maximum, d = minimum, c = inflection, b = slope

4.1. ELISA IgG and IgE titers

It was practical only to obtain sera from 4 out of the 8 members of the cohort. The results in Tables3 and4 are presented as the multiple of the mean of all control sera on all plates. This was chosen because it is the most conservative method of presentation for this type of data. Raw ELISA values ranged up to the 1.6384 E6 final dilution.

V0001 was immunized to all the venoms exceptN. annulata andD. polylepis. At the time of this blood sample, (2019) V0001 had begun inoculations withN. kaouthia venom. V0002 was immunized toC. atrox andB. atrox. For V0001 and V0002, other ELISA values are cross-reactive. V0007 immunized to all venoms shown, among others. V0008 was immunized only toA. squamigera,C. pyrrhus, andA. contortrix, which were not tested. Thus, all of V0008’s results were due to cross-reactivity.

Of the four sera tested, V0001 and V0007 displayed hyper-immune status by their tolerance to the venoms of species they were immunized to. V0001 did not attempt anN. kaouthia bite for another 9 months (≈12 more step cycles) and probably had much higher IgG titers at that time. It is not clear if V0002 and V0008’s titers shown would be sufficient to protect against a significant envenomation.

4.2. Adverse reactions to injections

Swelling (not shown inTable 5, shown inFigure 2) ranged from quite minor to involvement of an entire limb. Some degree of swelling occurs in virtually every injection with significant amounts of venom.

4.3. Dose determination and dry bites

The exact dose for a live bite is difficult to determine with precision. Bite envenomation doses are presented here as nominal wet venom with error bars. Doses are estimated by combining range data from literature with venom milking data and size of the snake. There was no evidence that any of these bites were prepped by pre-milking the snakes. The cohort reported that dry bites occurred multiple times that caused little pain and no swelling. Dry bite incidents were not reported for this study.

Figure 3’s bite envenomation events forNaja spp. are shown as an example.

4.4. Adverse bite envenomation events in context of immunizations

V0001’s first bite and infection (Table 6) was accidental and occurred at day 122 in his first immunization series. He sought medical attention at an ER immediately after seeking our advice. He was admitted for 2 days, given 12 vials of AV, and discharged. After discharge, he developed an antibiotic-resistant infection later, diagnosed as infection ofProteus mirabilis andEnterococcus spp. V0001 was readmitted for 21 days, receiving IV antibiotics, debridement, and distal finger joint amputation. V0001 is the only member of the cohort who had AV treatment for a bite. V0003’s bite (Table 6) was untreated.

The allergic events that occurred after V0007’s bites (Table 6) and also seen inFigure 4 consisted of severe hives, itching, and instances of throat swelling. With the exception of the throat involvement, symptoms appeared unexpectedly and did not recur.

4.5. Selected immunization series

Naive participants took a minimum of 90 days and a maximum of 18 months (Figure 5) to see doses of 10 mg or more with a minimum of inflammation. We take lack of inflammation as a signal of reasonably effective immunity. A number of venoms required considerably more time, on the order of 12–18 months, before sufficient immunity was seen. Exactly what drives this range is an open question that we will not speculate on.

Figure 4 shows high starting-dose immunizations conducted by an experienced member of the cohort (V0007) based on his assumption of cross-immunity. This participant had demonstrated effective hyperimmunity for other species within the elapids. For banded water cobra (N. annulata), wet venom dose started at 11.4 mg and progressed, within 60 days, to the chosen maintenance booster dose of 22.8 mg. Inflammation was a rough 3’’ × 3’’ region at this booster dose, which is considered tolerable. In contrast, the mamba (Dendroaspis spp.) species mixture resulted in anaphylactic reactions of varying intensities along with large inflamed regions. The participant did not suffer significant systemic effects, so cross-immunity was present and might have been mostly IgE based. What IgG cross-immunity existed was overwhelmed and the primacy of IgE is shown by the anaphylactic response. We make this diagnosis by examination of records. These anaphylactic reactions ceased abruptly around the 60 day mark, presumably because of development of sufficient IgG titers.

Figures4 and5, as well as Tables3 and4, provide a basis for the belief that a significant degree of cross-reactivity develops between snakes within the same family and greater cross-reactivity within the same genera. There is a great deal of conservation of sequence and structure in venom molecules, so cross-reactivity is to be expected. Mixtures of venoms are common to us for these vaccination series, and we believe that this is probably wise.

4.6. Pain of injections and bites by dose and elapsed days from starting injections

There was a good consistency in pain scale reports (Figure 6) across 6 different participants that provided pain data. One of the participants (V0007) captured a Northern paper wasp (Polistes fuscatus) and got it to sting several times. This was rated 2, which was consistent with the Starr scale [29]. The next day, a bite (5 bleeding fang holes) from a black mamba (D. polylepis) was rated 10 and declined to 8–9 (pain level) from h 1–3. In discussion with participants, the reason for the difference in subjective pain was ascribed to duration, intensity, and complexity. There are more tissues in a larger volume of the body that is affected by a snake bite than are affected by a typical hymenoptera sting.

Pain can be considerable in these inoculations, although some of the higher pain reports were associated with allergic reactions and infections. Once a booster dose is stabilized, the level of pain experienced from booster injections often remains high for years. Since pain has typically declined to a more manageable level within 2–3 h, there may be a place for physicians to provide local anesthetic for mixing with injections or for using before injecting venom.

4.7. Swelling by venom dose and elapsed days from starting injections

As seen inFigure 2, swelling declined dramatically over the course of 2–3 years of immunization. In the earliest period, some instances involved most of an arm or leg, and this occurred with Charles Tanner as well. Major swelling events might sometimes be associated with anaphylactic reactions and infections.

5.1. IgE has higher cross-immunity than IgG

IgE appeared to display greater cross-immunity between genera than IgG, and IgE response might last longer than IgG. So far, this has not been a problem except for overly enthusiastic immunization schedules for a new and distantly-related species.

5.2. Key relevant characteristics of venom

Venoms are pH-neutral or close to being pH-neutral, which means that pain is the result of venom components. Fresh, wet snake venom is approximately 60–83% water [32] with most venoms falling in the range of 70–80%. We used 75% as the nominal value in reconstitution and calculations when converting dry to wet form [33].

Wet venoms are extremely stable and have been stored for up to 80 years at room temperature in simple stoppered bottles, retaining their activity, even when some oxidation took place [34]. Desiccated neurotoxic venoms lost no activity in 8–9 years when stored in stoppered bottles in the dark, but hemorrhagic venoms lost roughly half in the same time period and roughly half again after 13 years. Desiccated venoms lost more than half of their toxicity in a year if the bottles were opened frequently [35]. This indicates that oxygen is the primary agent that depletes toxicity of desiccated venom and probably also of wet venom. However, injections with oxidized, deactivated, venom should be a kind of toxoid.

Study of field conditions also showed stability without refrigeration under adverse conditions [36]. Venoms contain antibacterial components to both Gram-negative and Gram-positive microorganisms [37].

Freeze-drying of venom has a significant effect on bacterial load, with greater killing of Gram-negative bacteria than their Gram-positive counterparts [38]. Freeze-drying without vacuum or storage in nitrogen exerts greater effect than using lyophilizing equipment. The freeze-drying procedure used by this cohort was to place wet venom in a vial with a cracked open top in an approximately −20°C freezer inside a jar packed with desiccant. While thisper se does not suffice to sterilize stored venom, and sterilization is not practical, it can be a helpful option. Venom components cause necrosis, neurotoxicity, myotoxicity, cardiotoxicity, hemorrhage, and thrombosis. PLA₂, serine proteases, and Zn²⁺-metalloproteases are the primary toxins of interest [39]. These can create an environment friendly to bacteria after injection.

Most practitioners keep supplies refrigerated or freeze dried. Concerns with whole venom instability over the time periods that most practitioners are likely to use them are minimal.

5.3. Kinetics of hyperimmunity

The kinetics of venom hyperimmunity should be the stoichiometry of the venom dose versus antibody present, both IgG and IgE [40]. Roughly half of IgG circulates in blood/lymph, the rest in interstitial spaces with interstitium circulation playing a major role [41]. In addition to elimination through kidneys, any cell that absorbs a venom molecule bound to antibody bearing an IgG Fc chain will bind TRIM21 and be ubiquitinylated for destruction by the proteasome and further presented to the immune system [42,43]. TRIM21 is a mechanism that is not available to most AVs due to lack of human Fc chain. We saw no evidence of antibody binding resulting in longer-term persistence of venom within our cohort.

Low molecular weight toxins, such as neurotoxins, diffuse rapidly, and high molecular weight toxins, such as cyto/myotoxins, spread slowly from the injection site [38]. Hence, neurotoxins and most hemotoxins will disseminate in the body and maximize their exposure to circulating antibodies. Conversely, to neutralize myotoxic/cytotoxic components, circulating antibodies have to be brought to the site.

For smaller venom molecules that permeate out into the body from the envenomation, this can be modeled simply against the total antibody to that antigen. For larger molecules, it needs to be modeled through transport of blood-bearing antibodies to the bite site. In practice, this can be a concern with bites to fingers.

5.4. Limits to hyper-immunity, polyclonality of antibodies, and overconfidence

As shown by the participant who required ICU care for 3 days with a flat EEG before regaining consciousness, there were limits to hyper-immunity. Our interpretation of this incident is that this person’s IgG was highly polyclonal because this person could accept 2 milking-style bites in quick succession without serious harm. It only required one antibody attaching to a protein (or virus) for the cell that encounters it to route that protein to the proteasome through the TRIM21 system. After a period of hours had passed from the first bite, most neurotoxic molecules of this first bite had soaked up at least double the number of antibodies required. This dramatically depleted the IgG antibodies available for the next bite. In the early phases of immunization, antibody levels are considerably lower than they can be later. We caution against overconfidence, particularly during the early stages.

5.5. Stoichiometric computation of hyperimmunity limits

The limits to hyperimmunity in humans have not yet been determined experimentally. It is also unknown what the practical limits of antibody generation are in humans, and these limits probably vary significantly. Here, we present guideline estimates.

The current limit understanding is presented here through an example based on a large specimen of Gaboon viper,Bitis gabonica. This kind of bite is probably beyond the ability of immunization to fully neutralize, except possibly for a human of approximately 440 lbs (200 kg).

A single milking of a Gaboon viper yielded as much as 2 g of dried venom [44]. Venom is 70–80% water. That implies 6–8 mL of wet venom. Average yield is 200–1000 mg of dry venom.

We will assume 1000 mg of dry venom, which corresponds to an approximate 4 mL bite from a large snake, assuming 75% water in the venom. We will use an average venom molecular weight of 13,000 g (13 kDa).

1 g of venom ÷ 13,000 g/mole ≈ 7.69E-5 moles of venom (76.9 micromoles)

7.69E-5 · 6.0221409E23 ≈ 4.63E19 molecules of venom.

A 70 kg adult human male has about 5 L of blood, with roughly 3 L of plasma/lymph, mostly in interstitial tissues. The plasma/lymph has a higher ratio of antibodies because it is just serum. Plasma makes up approximately 55% of blood volume so the 3 L of plasma/lymph in the body should have an equivalent load of IgG given by:

3 L ÷ 0.55 ≈ 5.45. This gives us a total of roughly 10.45 L (104.5 dL) of blood equivalent antibody-bearing fluid, in a 70 kg adult human male.

Normal total IgG levels average about 1000 mg per dL, with a normal adult range from 639 to 1349 mg of IgG per dL [45]. This allows for the calculation of a range.

In animals used for the production of antivenin, the best estimate of IgG fraction specific to venom is 1–2% [46]. We corresponded on the ELISA results from V0007 [47] who has hyper-immunized himself to several snakes, and if those were correct, V0007 may have roughly doubled the normal level. But even with that, V0007 should not have more than about 15% of that possible doubled IgG devoted to one snake species, in the most generous estimate. Cross-reactive antibodies are possible, and there is definitely evidence of that, but we will ignore it for this example.

Table 7 indicates a range from about 9–42 micromoles within normal human range, and inTable 8, maximum range may be as high as roughly 210–630 micromoles of specific IgG to a venom. The values inTable 8 are probably overestimates, but we cannot rule them out at this time. A fairly large Gaboon viper would inject approximately 77 micromoles of toxins. A very large one might inject 150 micromoles. It is evident that, within the normal range, such a snake should overwhelm antibody-based immunity, although it will be helpful. However, the problem is more complicated than that.

Working through the problem of calculating the limits to hyperimmunity is not as simple as stoichiometric balancing as if the human body was a beaker of venom and antibody. Antibody is dispersed throughout the bloodstream and interstitial plasma/lymph. The interstitial plasma turns over about once every 24 h normally. To better understand, this would require a good simulation model and comparison with experimental subjects.

For hyperimmunity to function, there must be perfusion of antibody to the venom molecules, to allow for binding. Binding is followed by ingestion by a cell so that the protein toxin is destroyed or eliminated by the kidneys.

Different venom components will have different requirements to neutralize them. The easiest are the neurotoxins because they must perfuse through the body to act, which puts them into contact with antibody. Second to these are the hemotoxins which have overwhelming effect at the site of the bite and then travel through the blood stream, counteracting clotting mechanisms. Perhaps, the most problematic for hyperimmunity are the cytotoxins which move slowly in the body and stay concentrated, causing tissue damage.

5.6. Concerns regarding autoimmunity induction were not seen in or outside this cohort

We tried to address the issue of possible generation of auto-immunity from venom immunization by examining this cohort and searching for reports of long-term health issues in those outside the cohort. While the sample was small, no suggestions of auto-immunity being an issue have shown up. Practitioners arrived at their second half-century in excellent health. Several long-term practitioners outside the dataset, on the order of 100 man-years, showed no evidence of auto-immunity, and we speculate that this practice might be protective. Bill Haast, with 64 man-years of practice, was lively and active until near his death at the age of 100, without noticeable arthritis or anything else that was auto-immunity related. Granted, Mr. Haast injected lower doses than most in this cohort, and he suffered from serious damage to his hands. In animal studies of auto-immunity, Freund’s complete adjuvant is typically used with conjugated self-proteins which venom injections do not have.

5.7. Cautionary notes regarding public claims of venom immunization

We established that not all who claim to be immunizing with venoms actually are including a prominent proponent. Careful evaluation and patience may result in admission that either immunization did not occur at all or it was greatly exaggerated after an original intent to do so. The pain induced by these injections, even from a miniscule dose, is no laughing matter. People vary greatly in their pain threshold and their ability to handle pain. When ELISA results do not match claims, or if there is a veering response to requests for corroborating data, or data record sheets look “too good,” scientists need to be careful about accepting claims at face value. A sympathetic approach is needed to this kind of investigation on the human side that many laboratory scientists are unfamiliar with but should be a quite familiar territory to healthcare workers and psychologists. A person may find themselves painted into a corner by their initial promises.

5.8. Risks of injecting with snake venoms

5.9. Gastão Rosenfeld’s rejection of immunization to snake venoms

Rosenfeld stated a view about venom immunization that is widely held in herpetology and the venom studies field [31]:

• “Active immunization of man was tried with success (Haast and Winer, 1955; Wiener, 1960; Flowers, 1963), but an a priori affirmation can be expressed that it is not satisfactory due to several factors. Instituto Butantan employees working at the serpentarium have been bitten repeatedly and showed no immunity to subsequent bites. Horses immunized for antivenin production have a shorter life-span than normal. Necropsy [animals] showed tissue degeneration of organs, such as the turbid swelling of the liver (Vaz and Araujo, 1948) probably as a result of the repeated venom injections needed to maintain high antibody count.

• The injection of small and repeated doses of venom induces a good level of antibody formation. Soon after stopping, antibody titer decreases and comes almost to zero. When a new immunization series is started in these animals, the high antibody titer is reached sooner than the first time, but this state will take days or weeks.” - Gastão Rosenfeld 1971

Rosenfeld’s accomplishments were seminal, and he laid the groundwork for the development of angiotensin-converting enzyme-inhibiting drugs. He deserves the respect he commands in the fields of herpetology and venom studies. However, in light of more current understanding, we can address the objections he raised to venom immunization in humans.

As above mentioned, Rosenfeld stated that immunization in humans was, indeed, successful. Charles Tanner was immunized toNotechis scutatus (Tiger snake), and Charles tested with 25 mg of wet venom [8]. Charles Tanner showed that his immunity persisted for 3 months, and hisN. scutatus bite recovery without AV was uneventful [8]. Just 1 mL of serum from Flowers neutralized 30 mouse LD50’s, and Flowers’ serum should have been able to easily neutralize 100 mg ofNaja naja (Indian cobra) venom [9]. Bill Haast survived aBungarus candidus (Blue krait) bite without AV [7]. A smallBungarus multicinctus (Multi-banded krait) is what killed Joseph Slowinski from a bite that did not appear to break the skin [16].

Taking Rosenfeld’s first point, envenomated employees would be treated, which blunts B-cell response, and even without treatment, stimulations at long intervals will not produce hyper-immunity, which is not a normal state for human immune systems. Both points were made by Flowers in the second paragraph of his paper [9]. That said, based on snakebite symptomatic response, one member of the VIPRBITEM cohort did appear to exhibit hyper-immune status primarily from receiving regular snakebites. Unfortunately, this person drowned accidentally a week before serum could be acquired.

To Rosenfeld’s second point about life span of horses used for AV production, the dose escalation schedule used in AV production is far more aggressive than anything used on humans. Based on this study, it would take 12–18 months for an animal to comfortably develop hyperimmune status for snake venoms. However, there may be quirks in different animal species that change the schedule and upper limit somewhat. Members of this human cohort increased doses slowly, except for one who increased rapidly because of error and had to stop. As Rosenfeld mentioned above, liver and kidney damage in animals occurs from overly rapid dose escalation. In the VIPRBITEM cohort, blood work showed no indication of harm to the liver or kidneys. If antibody levels rise so as to neutralize venom effectively, harm to liver and kidneys is not expected.

Further addressing Rosenfeld’s second point about life span of animals, which implies cutting life span of humans, he could not have known in 1972 that Bill Haast would live to be 100 (22 years longer than Gastão). Bill continued his weekly practice of injections until shortly before his death in 2011, spanning 64 years. We do not see any indications within the cohort that venom vaccination injections cause harm that could shorten life span. The first author is very sensitive to this because he works on increasing health span and potential life span in humans.

In the VIPRBITEM cohort, the longest interval that anyone has gone between a booster and getting bitten without needing treatment by AV was nearly a year. This individual has a two-decade history of injections and bites and extraordinary antibody titers to venoms. Our data indicated that boosters every 3 months once, hyperimmunity has been established are probably adequate, and there may be room to increase the interval. We do not know, at this time, what the optimal booster interval is, nor how human hyper-immunity behaves over time with precision. This warrants more studies. However, it is quite clear, from our ELISA results, that hyper-immune status does not rapidly return to zero as Rosenfeld described. Without numbers attached to what Rosenfeld thought a small venom dose was, and without defining what a “good level of antibody” means, we can only presume that these doses were very small, as was the antibody response, and that Rosenfeld’s experiments bore little resemblance to the schedule presented in Weiner’s paper on Charles Tanner [8].

• 1.Lucan P, Cambridge MA. Harvard University Press. Circa 60 AD. Available from:https://archive.org/stream/lucancivilwarboo00lucauoft/lucancivilwarboo00lucauoft_djvu/txt. [Last accessed on 2018 Jan 10].
• 2.Dio C. Dio's Roman History Vol LI.14:Google Digital Books. Circa 210 AD. [Google Scholar]
• 3.Pinto S. How I Crossed Africa:From the Atlantic to the Indian Ocean, through Unknown Countries;Discovery of the Great Zambezi Affluents Vol 2. London:Sampson Low, Marston, Searle and Rivington;1881. [Google Scholar]
• 4.Calmette A. Venoms:Venomous Animals and Antivenomous Serum-therapeutics. London:John Bale and Sons, and Danielsson, Ltd;1908. [Google Scholar]
• 5.Carnochan FG, Adamson HC. Empire of the Snakes. London:Hutchinson and Co.;1935. [Google Scholar]
• 6.Webb GB, Powell D. Henry Sewall Physiologist and Physician. Baltimore:The John Hopkins Press;1946. [Google Scholar]
• 7.Haast WE, Winer ML. Complete and spontaneous recovery from the bite of a blue krait snake (Bungarus caeruleus). Am J Trop Med Hyg. 1955;4(6)1135-1137. 10.4269/ajtmh.1955.4.1135. [DOI] [PubMed] [Google Scholar]
• 8.Wiener S. Active immunization of man against the venom of the Australian tiger snake (Notechis scutatus). Am J Trop Med Hyg. 1960;9(3)284-292. 10.4269/ajtmh.1960.9.284. [DOI] [PubMed] [Google Scholar]
• 9.Flowers HH. Active immunization of a human being against Cobra (Naja naja) venom. Nature. 1963;200:1017-1018. 10.1038/2001017b0. [DOI] [PubMed] [Google Scholar]
• 10.Canan ED, Flowers HH. Cobra bite following immunization against cobra venom. JAMA. 1965;193:625-626. 10.1001/jama.1965.03090070075034. [DOI] [PubMed] [Google Scholar]
• 11.Kursh H. Cobras in His Garden. Goa:Harvey House;1965. [Google Scholar]
• 12.Klauber LM. Rattlesnakes:Their Habits, Life Histories and Influence on Mankind. Oakland, CA:University of California Press;1979. [Google Scholar]
• 13.Haast N. The official Web Site of Bill Haast:Snakebites and Immunization Miami. Florida:Miami Serpentarium Laboratories; Available from:https://www.billhaast.com/serpentarium/immunization_snakebites.html. [Last accessed on 2017 May 20]. [Google Scholar]
• 14.Israel BA, Schulz AJ, Parker EA, Becker AB. Community-based participatory research:Policy recommendations for promoting a partnership approach in health research. Educ Health (Abingdon). 2001;14(2)182-197. 10.1080/13576280110051055. [DOI] [PubMed] [Google Scholar]
• 15.Ball MP, Bobe JR, Chou MF, et al. Harvard personal genome project:Lessons from participatory public research. Genome Med. 2014;6(2. 10.1186/gm52. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 16.Moffett MW. Bit. Outside Online. Boulder:Outside;2002.http://outside.away.com/outside/adventure/200204/200204_bit_1.html. [Google Scholar]
• 17.Boyer LV. On 1000-fold pharmaceutical price markups and why drugs cost more in the United States than in Mexico. Am J Med. 2015;128(12)1265-1267. 10.1016/j.amjmed.2015.08.007. [DOI] [PubMed] [Google Scholar]
• 18.Joseph Pittman RN. North American Pit Viper Bite:Ranges of Vial Use for Envenomation. Personal Communication. Washington DC:CSPI;2017. [Google Scholar]
• 19.Himmelstein DU, Thorne D, Warren E, Woolhandler S. Medical bankruptcy in the United States, 2007:Results of a national study. Am J Med. 2009;122(8)741-746. 10.1016/j.amjmed.2009.04.012. [DOI] [PubMed] [Google Scholar]
• 20.McGhee S, Finnegan A, Clochesy JM, Visovsky C. Effects of snake envenomation:A guide for emergency nurses. Emerg Nurse. 2015;22(9)24-29. 10.7748/en.22.9.24.e1406. [DOI] [PubMed] [Google Scholar]
• 21.Spano SJ, Vohra R, Macias F. Long-term complications of rattlesnake bites:A telephone survey from Central California. Wilderness Environ Med. 2014;25(2)210-213. 10.1016/j.wem.2013.11.004. [DOI] [PubMed] [Google Scholar]
• 22.Benoit N. Links. Florida:Norman Benoit;2005. Available from:https://web.archive.org/web/20111115174034;https://normanbenoit.com/links.htm. [Last accessed on 2017 Jun 23]. [Google Scholar]
• 23.Brown SG, Wiese MD, Van Eeden P, et al. Ultrarush versus semirush initiation of insect venom immunotherapy:A randomized controlled trial. J Allergy Clin Immunol. 2012;130(1)162-168. 10.1016/j.jaci.2012.02.022. [DOI] [PubMed] [Google Scholar]
• 24.Hanley BP, Gross G. Extraordinary variance in meta-analysis of venom toxicity of 160 most lethal ophidians, and guidelines for estimating human lethal dose range. J Biol Methods. 2024;11(3)e99010029. 10.14440/jbm.2024.0037. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 25.Hanly WC, Artwohl JE, Bennett BT. Review of polyclonal antibody production procedures in mammals and poultry. ILAR J. 1995;37(3)93-118. 10.1093/ilar.37.3.93. [DOI] [PubMed] [Google Scholar]
• 26.Johnson WE, Oyervides M, Forstner MR. Drymarchon melanurus erebennus (Texas Indigo Snake) Defensive behavior/death-feigning. Herpetol Rev. 2017;48(2)448. [Google Scholar]
• 27.Chen CM, Wu KG, Chen CJ, Wang CM. Bacterial infection in association with snakebite:A 10-year experience in a Northern Taiwan medical center. J Microbiol Immunol Infect. 2011;44(6)456-460. 10.1016/j.jmii.2011.04.011. [DOI] [PubMed] [Google Scholar]
• 28.LoVecchio F, Klemens J, Welch S, Rodriguez R. Antibiotics after rattlesnake envenomation. J Emerg Med. 2002;23(4)327-328. 10.1016/s0736-4679(02)00563-2. [DOI] [PubMed] [Google Scholar]
• 29.Starr CK. A simple pain scale for field comparison of hymenopteran stings. J Entomol Sci. 1985;20(2)225-231. 10.18474/0749-8004-20.2.225. [DOI] [Google Scholar]
• 30.Wiener S. Snake bite in a subject actively immunized against snake venom. Med J Aust. 1961;48(1)658-659. 10.5694/j.1326-5377.1961.tb69017.x. [DOI] [PubMed] [Google Scholar]
• 31.Rosenfeld G. Symptomatology, pathology, and treatment of snake bites in South America Bucherl W, Buckley EE, Deulofeu V. Venomous Animals and their Venoms2;New York:Academic Press;1971:380. [Google Scholar]
• 32.Abdel-Aal A, Abdel-Baset A. Venom yield and toxicities of six Egyptian snakes with a description of a procedure for estimating the amount of venom ejected by a single snake bite. Sci J King Faisal Univ (Basic Appl Sci). 2010;11(1)167-182. [Google Scholar]
• 33.Mirtschin PJ, Dunstan N, Hough B, et al. Venom yields from Australian and some other species of snakes. Ecotoxicology. 2006;15(6)531-538. 10.1007/s10646-006-0089-x. [DOI] [PubMed] [Google Scholar]
• 34.Jesupret C, Baumann K, Jackson TN, et al. Vintage venoms:Proteomic and pharmacological stability of snake venoms stored for up to eight decades. J Proteomics. 2014;105:285-294. 10.1016/j.jprot.2014.01.004. [DOI] [PubMed] [Google Scholar]
• 35.Schöttler WH. On the stability of desiccated snake venoms. J Immunol. 1951;67(4)299. 10.4049/jimmunol.67.4.299. [DOI] [PubMed] [Google Scholar]
• 36.Munekiyo SM, Mackessy SP. Effects of temperature and storage conditions on the electrophoretic, toxic and enzymatic stability of venom components. Comp Biochem Physiol Part B Biochem Mol Biol. 1998;119(1)119-127. 10.1016/s0305-0491(97)00294-0. [DOI] [PubMed] [Google Scholar]
• 37.Perumal Samy R, Gopalakrishnakone P, Thwin MM, et al. Antibacterial activity of snake, scorpion and bee venoms:A comparison with purified venom phospholipase A2 enzymes. J Appl Microbiol. 2007;102(3)650-659. 10.1111/j.1365-2672.2006.03161.x. [DOI] [PubMed] [Google Scholar]
• 38.Carvalho AS, Silva J, Ho P, Teixeira P, Malcata FX, Gibbs P. Effect of various growth media upon survival during storage of freeze-dried Enterococcus faecalis and Enterococcus durans. J Appl Microbiol. 2003;94(6)947-952. 10.1046/j.1365-2672.2003.01853.x. [DOI] [PubMed] [Google Scholar]
• 39.Calvete JJ, Sanz L, Angulo Y, Lomonte B, Gutierrez JM. Venoms, venomics, antivenomics. FEBS Lett. 2009;583(11)1736-1743. 10.1016/j.febslet.2009.03.029. [DOI] [PubMed] [Google Scholar]
• 40.Starkl P, Marichal T, Gaudenzio N, et al. IgE antibodies, FceRIa, and IgE-mediated local anaphylaxis can limit snake venom toxicity. J Allergy Clin Immunol. 2016;137(1)246-257.e211. 10.1016/j.jaci.2015.08.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 41.Gitlin D. Some concepts of plasma protein metabolism, A. D. 1956. Pediatrics. 1957;19(4)657. [PubMed] [Google Scholar]
• 42.Mallery DL, McEwan WA, Bidgood SR, Towers GJ, Johnson CM, James LC. Antibodies mediate intracellular immunity through tripartite motif-containing 21 (TRIM21). Proc Natil Acad Sci United States Am. 2010;107(46)19985-19990. 10.1073/pnas.1014074107. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 43.McEwan WA, Tam JC, Watkinson RE, Bidgood SR, Mallery DL, James LC. Intracellular antibody-bound pathogens stimulate immune signaling via Fc-receptor TRIM21. Nat Immunol. 2013;14(4)327-336. 10.1038/ni.2548. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 44.Francischetti IM, My-Pham V, Harrison J, Garfield MK, Ribeiro JM. Bitis gabonica (Gaboon viper) snake venom gland:Toward a catalog for the full-length transcripts (cDNA) and proteins. Gene. 2004;337:55-69. 10.1016/j.gene.2004.03.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 45.Agarwal S, Cunningham-Rundles C. Assessment and clinical interpretation of reduced IgG values. Ann Allergy Asthma Immunol. 2007;99(3)281-283. 10.1016/s1081-1206(10)60665-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 46.Boyer L. Fraction of IgG in Horses Specific to one Venom's Components. Personal Communication to Brian Hanley. 2015. [Google Scholar]
• 47.Kagen S, Muthiah R. Poisonous snake venom vaccination:A case report of a non-physician directed experiment. Ann Allergy Asthma Immunol. 2004;92(1)120. [Google Scholar]
• 48.Langley RL. A review of venomous animal bites and stings in pregnant patients. Wilderness Environ Med. 2004;15(3)207-215. 10.1580/1080-6032(2004)15[207:arovab]2.0.co;2. [DOI] [PubMed] [Google Scholar]
• 49.Palmeira P, Quinello C, Silveira-Lessa AL, Zago CA, Carneiro-Sampaio M. IgG placental transfer in healthy and pathological pregnancies. Clin Dev Immunol. 2012;2012:985646. 10.1155/2012/985646. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 50.Dehghani R, Sharif MR, Moniri R, Sharif A, Kashani HH. The identification of bacterial flora in oral cavity of snakes. Comp Clin Pathol. 2016;25(2)279-283. 10.1007/s00580-015-2178-9. [DOI] [Google Scholar]
• 51.Goldstein EJ, Citron DM, Gonzalez H, Russell FE, Finegold SM. Bacteriology of rattlesnake venom and implications for therapy. J Infect Dis. 1979;140(5)818-821. 10.1093/infdis/140.5.818. [DOI] [PubMed] [Google Scholar]
• 52.Jho YS, Park DH, Lee JH, Cha SY, Han JS. Identification of bacteria from the oral cavity and cloaca of snakes imported from Vietnam. Lab Anim Res. 2011;27(3)213-217. 10.5625/lar.2011.27.3.213. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 53.Theakston RD, Phillips RE, Looareesuwan S, Echeverria P, Makin T, Warrell DA. Bacteriological studies of the venom and mouth cavities of wild Malayan pit vipers (Calloselasma rhodostoma) in southern Thailand. Trans R Soc Trop Med Hyg. 1990;84(6)875-879. 10.1016/0035-9203(90)90112-r. [DOI] [PubMed] [Google Scholar]
• 54.Sriprapat S, Aeksowan S, Sapsutthipas S, et al. The impact of a low dose, low volume, multi-site immunization on the production of therapeutic antivenoms in Thailand. Toxicon. 2003;41(1)57-64. 10.1016/s0041-0101(02)00209-x. [DOI] [PubMed] [Google Scholar]
• 55.Klinkenberg J. Florida's Snake Man Made Living from Deadly Serpents. United States:Tampa Bay Times;2011. [Google Scholar]

Data Availability Statement

Data used in this work is available from the corresponding author upon reasonable request.