An updated review of feline coronavirus: mind the two biotypes

Highlights

• •The pathogenicities of the two biotypes of feline coronavirus differ.
• •Internal mutation occurs in the two biotypes of feline coronavirus.
• •Effusive and noneffusive FIP can be transformed mutually based on immune status.

Abstract

Feline coronavirus (FCoV) includes two biotypes: feline infectious peritonitis virus (FIPV) and feline enteric coronavirus (FECV). Although both biotypes can infect cats, their pathogenicities differ. The FIPV biotype is more virulent than the FECV biotype and can cause peritonitis or even death in cats, while most FECV biotypes do not cause lesions. Even pathogenic strains of the FECV biotype can cause only mild enteritis because of their very low virulence. This article reviews recent progress in FCoV research with regard to FCoV etiological characteristics; epidemiology; clinical symptoms and pathological changes; pathogenesis; and current diagnosis, prevention and treatment methods. It is hoped that this review will provide a reference for further research on FCoV and other coronaviruses.

Graphical abstract

1. Introduction

Feline coronavirus (FCoV) includes two biotypes: feline infectious peritonitis virus (FIPV) and feline enteric coronavirus (FECV). Infectious peritonitis in cats was first reported in 1966 by Wolfe and Griesemer (1966), who introduced feline infectious peritonitis (FIP) as a disease; however, because of the conditions at the time, only bacteriological tests were performed to detect bacterial infections. It was not until 1970 that Ward (1970) first isolated a mouse hepatitis virus (MHV)-like virus in cats with peritonitis, which was unlike the previously isolated feline leukemia virus (FeLV). Ward and coworkers found that the particles of the new virus were smaller in diameter than FeLV particles and had many spikes distributed on their surfaces, similar to MHV particles. In 1978, Pedersen et al. (1978) compared the antigenicity of FIPV with that of seven other coronaviruses by direct and indirect immunofluorescence methods and found that FIPV was very closely related to transmissible gastroenteritis virus (TGEV) and MHV3, which belong to the family Coronaviridae; therefore, FIPV was officially classified as a coronavirus. However, it was found during the discovery, naming and classification of FIPV that the virus cannot simply be equated with FIP because infectious peritonitis is only one of the pathological changes caused by FIPV in cats. In this sense, the name “FIPV” is not accurate, but it has become cemented in the lexicon through continuous use. FECV was discovered in 1981 by Pedersen et al. (1981). This virus is similar to FIPV in its antigenicity but different in its pathogenicity. FIPV can cause systemic infection in animals, while FECV usually leads to mild intestinal symptoms. Therefore, FIPV is more virulent than FECV. FCoV is described in detail below.

2. Etiological characteristics

FCoV is a nonsegmented, single-stranded RNA virus that belongs to the order Nidovirales, family Coronaviridae, subfamily Coronavirinae and genus Alphacoronavirus 1. The virus particles are round or polymorphic, with helical symmetry and diameters of approximately 80–120 nm. FCoV is enveloped and has approximately 15–20-nm-long spikes on its surface, which cause a coronal (crownlike) appearance under an electron microscope (Jaimes and Whittaker, 2018) (Fig. 1a).

Fig. 1.

(a) Schematic representation of FCoV (Jaimes and Whittaker, 2018). (b) Schematic representation of the genomes of FCoV I and FCoV II (Gao et al., 2022; Le Poder, 2011).

Note: The genomes of FCoV I and FCoV II (represented by FIPV-II-79–1146 in this article) are both approximately 29 kb in size and contain 11 ORFs encoding four structural proteins (S, M, E and N) and seven nonstructural proteins (replicase proteins 1a and 1b and accessory proteins 3a, 3b, 3c, 7a and 7b). Light blue indicates the genes from FCoV I, while dark blue indicates the genes from canine coronavirus (CCoV). The S protein consists of two subunits: S1 (the receptor binding domain, RBD) and S2 (the fusion domain, FD). The S1 subunit also includes two functional domains: the amino terminal domain (NTD) and the carboxyl terminal domain (CTD). The S2 subunit consists of a fusion peptide (FP), two heptad repeats (HR1 and HR2), a transmembrane (TM) domain and an internal domain (ID). The arrows indicate the two activation sites, S1/S2 and S2′, and the linker sequence (L) region between S1/S2 and S2′. The upper dotted line in each FCoV genome representation indicates the difference in the key amino acid 1058 of the S gene between FECV and FIPV, and the lower dotted line indicates the difference in the S1/S2 furin cleavage site between FECV and FIPV.

As shown in Fig. 1b, the genome of the virus, which is approximately 29 kb in size, includes 11 open reading frames (ORFs) and encodes four major structural proteins: the spike (S), nucleocapsid (N), envelope (E) and membrane (M) proteins (de Barros et al., 2019). It also encodes seven nonstructural proteins, including two replicase proteins (1a and 1b) and five accessory proteins (3a, 3b, 3c, 7a and 7b) (Gao et al., 2022). FCoV is divided into two genotypes according to its antigenicity, FCoV I and FCoV II, and each of them can be divided into two biotypes based on its pathogenicity: FECV and FIPV (Decaro et al., 2021). The representative strain of FCoV I is FECV-UCD, also named FECV-I-UCD, which was isolated by Pedersen et al. (1981). Other FCoV I strains, including FIPV-I-UCD1, FIPV-I-UCD3, FIPV-I-UCD4 and FIPV-I-TN406, are shown in Table 1 (Motokawa et al., 1996; Pedersen et al., 1981). FCoV I is difficult to reproduce in cell culture, as it grows well only in macrophages (Pedersen et al., 1981). It also produces a low virus yield and has a weak specific serum response to canine coronavirus (CCV). The representative strains of FCoV II are FIPV-II-NOR 15, which was isolated by Evermann et al. (1981), and FIPV-II-1683 and FIPV-II-1146, which were isolated by Pedersen and coworkers (Pedersen et al., 1984). Detailed information on these FCoV II strains is also shown in Table 1. It has been reported that FCoV II originated by homologous recombination between FCoV I and CCV (Herrewegh et al., 1998) and that it can proliferate in many cell lines, produce a high virus yield and react strongly with CCV antiserum. FCoV I is more widespread in terms of natural infections worldwide, while FCoV II infection is less common. However, most studies on FCoV thus far have focused on FCoV II, while there have been few studies on the very common FCoV I in cell cultures (Amer et al., 2012). Previous studies have shown that FCoV I and FCoV II have different receptors. Feline aminopeptidase-N (fAPN) has been shown to act as a receptor for FCoV II but not for FCoV I. It has been reported that feline dendritic cell-specific intercellular adhesion molecule grabbing nonintegrin (fDC-SIGN) is an important virus binding receptor for FIPV I but not for FIPV II, because the fDC-SIGN is responsible for the FIPV I binding but not for FIPV binding, although both serotype I and II FIPVs use fDC-SIGN for infection of monocyte-derived cells (Gao et al., 2022; Van Hamme et al., 2011). Importantly, both FCoV I and FCoV II can be categorized as FECV and FIPV depending on observation. The two biotypes of FIPV and FECV differ substantially in their biological characteristics (Tekes and Thiel, 2016). FIPV is more virulent than FECV and can even cause death in cats within 1–3 weeks after infection (Wolfe and Griesemer, 1966), while FECV usually causes only mild intestinal symptoms (Desmarets et al., 2016). In addition to differences in pathogenicity in vivo, the two biotypes also display different tropisms (Pedersen, 2009). In general, FECV proliferation is confined to intestinal mucosal epithelial cells or mesenteric lymph nodes during natural infection (Pedersen et al., 1981; Stoddart et al., 1988), whereas FIPV infects feline peritoneal macrophages and can then infect more types of cells, including monocytes, plasma cells, lymphocytes and neurocytes (Pedersen, 2009), resulting in high titers of progeny virus. FIPV can thus infect other types of cells continuously and more efficiently than FECV (Dewerchin et al., 2005; Stoddart and Scott, 1989).

Table 1.

Information on different FCoV strains.

Primary name	Serotype	Biotype	Nomenclature	NCBI accession number
FECV-UCD	I	FECV	FECV-I-UCD	X90574
FIPV-UCD1	I	FIPV	FIPV-I-UCD1	X90575
FIPV-UCD3	I	FIPV	FIPV-I-UCD3	X90577
FIPV-UCD4	I	FIPV	FIPV-I-UCD4	X90578
FIPV-TN406	I	FIPV	FIPV-I-TN406	X90570
FIPV-NOR 15	II	FIPV	FIPV-II-NOR 15	MW225969
FCoV 79–1683	II	FECV	FEPV-II-79–183	JN634064
FCoV 79–1146	II	FIPV	FIPV-II-79–1146	DQ010921

3. Epidemiology of FCoV

3.1. FIPV biotype

FIP is endemic worldwide, and FIPV can infect domestic and wild cats of all ages. However, the highest incidence of FIP occurs in cats aged 3 months to 2 years, followed by cats over 10 years old (Rohrbach et al., 2001). Studies have shown that male and unsterilized cats are more susceptible to FIP than female and sterilized cats, suggesting that the disease may be associated with sex (Worthing et al., 2012). Some scholars have proposed that purebred cats are more susceptible to FIP than crossbred cats (Kiss et al., 2000), but no scientific basis has been found. Interestingly, the incidence in British Shorthair cats is higher than that in other cat breeds (Worthing et al., 2012). High breeding density and an unsanitary breeding environment are closely related to the occurrence of FIP. In addition to domestic cats, FIPV can also infect African lions, mountain lions, leopards, cheetahs, jaguars, lynx, servals, caracals, European wildcats, sand cats, Pallas’ cats and nearly all other catamounts (Pedersen, 2009).

The incidence of this disease is difficult to determine and differs among countries and regions. Approximately 1–5% of cats from catteries and shelters died of FIP in the United States from 1963 to 2008 (Pedersen, 2009). In China, Li collected 115 samples of ascetic fluid from FIP-suspected cats and 54 fecal samples from clinically healthy cats from veterinary hospitals to detect the presence of FCoV in the samples by RT-PCR recently. The results showed that a total of 126 (74.6%, 126/169) samples were positive for FCoV (Li et al., 2019). Notably, the occurrence of FIP is closely related to the occurrence of FECV; when FECV occurs, FIP also occurs (Pedersen, 2009). Regarding the serotypes of FcoV, FCoV I infection is more common than FCoV II infection, with an incidence of 80–95% in Europe and the United States, whereas FCoV II infection is rarely reported (Benetka et al., 2004; Kummrow et al., 2005). In Asia, however, the incidence of FCoV II has been found to reach 2.5–29.5% (Amer et al., 2012; Zhou et al., 2021b).

3.2. FECV biotype

Virions of FECV are shed mainly in feces, and healthy cats can be infected through contact with contaminated environments. Pederson et al. reported that FECV shedding increases by 10- to 10⁶-fold within a week after stray cats are placed in shelters in the United States (Pedersen et al., 2004). It has been reported that 6- to 12-week-old kittens are the most susceptible to infection, but the age with the highest susceptibility under laboratory conditions is approximately 12–15 weeks; this inconsistency may be due to differences in the environment and feeding conditions (Pedersen et al., 2008). Kittens are protected by antibodies, such as IgG and IgA, obtained from their mother's colostrum (Pearson et al., 2019); kittens over 5 weeks of age are susceptible to the virus because their antibody titers are decreased. Six- to twelve-week-old cats often show symptoms of gastroenteritis, while adult cats are generally asymptomatic, but sometimes cases can be serious (Pedersen et al., 2008, 1981). Once infected, sick cats and healthy cats carrying the virus shed large numbers of virus particles into the environment through their feces. There are three patterns of virus particle shedding: persistent shedding, in which particles are shed at varying levels over 9–24 months; shedding interlaced with periods of nonshedding; and shedding that ceases after 5–19 months (average of 12 months) (Pedersen et al., 2008). Notably, most recovered cats still carry FECV. Pedersen et al. (2008) showed that in FECV-infected cats, a higher antibody titer is associated with a greater ability to shed the virus into the environment. In addition, cats that have recovered from FECV infection can still become infected upon virus challenge, which might be related to a common feature of the immune response to enteric coronaviruses (Brandtzaeg, 2007; Pedersen et al., 2008; Saif, 2004).

The FECV carrier rate is closely related to the life habits and immune statuses of animals. For example, studies have shown that gregarious cats have a higher infection rate than solitary cats (McKay et al., 2020; Pedersen, 2009; Pedersen et al., 2004). In addition, a survey by Cave et al. of more than 2000 cats from 14 stray cat rescue shelters in the UK showed that the risk of infection is 2.3 times higher in cats kept in groups than in cats kept alone (Cave et al., 2004).

4. Clinical symptoms and pathological changes in FCoV infection

4.1. FIPV biotype

FIPV infection in cats generally produces the following clinical symptoms: lassitude, inappetence, dehydration, diarrhea, and sometimes undulating fever. Once the body temperature of an infected cat exceeds 39°C, weight loss and increased urination occur, and some cats may develop jaundice (Sykes, 2013). According to the disease characteristics, FIP can be divided into two types: effusive and noneffusive (Addie et al., 2020a).

Effusive-type FIP is also known as wet peritonitis, and its main symptom is abdominal distension featuring abdominal swelling with a positive fluid wave upon palpation but no pain. Female cats are often misdiagnosed with pregnancy or weight gain. Male cats that are intact may have scrotal swelling. Abdominal masses may be detected on palpation, indicating organ–omentum adhesion. Moreover, the mesenteric lymph nodes may be enlarged. Abdominal effusion fluid is generally yellow, clear or cloudy and mucinous. If a cat has pleural effusion, it may also present with dyspnea and lung noise on auscultation (Pedersen, 2009). Some cats develop pericardial effusion. Icterus may occur in the terminal stage of the disease (Sykes, 2013). In addition, a small number of cats with effusive peritonitis will develop ocular disease and central nervous system disease.

Noneffusive FIP, also known as dry FIP, has a chronic progressive course. It usually takes weeks or months after the first infection for a cat to develop symptoms of the disease. The characteristic lesions of dry FIP are granulomas, which are distributed mainly in the abdominal organs, including the liver, kidneys, mesenteric lymph nodes, and especially the T-shaped junction of the colon and ileocecal valve. Approximately 60% of cats with noneffusive FIP suffer from ocular and central nervous system damage, and approximately 40% have abdominal lesions, which may or may not be associated with ocular and central nervous system lesions. The main manifestations of ocular lesions are uveitis and chorioretinitis. First, the color of the iris of the eyeball changes; this change is accompanied by hyperemia and the presence of precipitates. The iris then becomes white or colorless. At this time, corneal opacity and precipitates will be present, and the shape of the pupils is sometimes distorted. In some cases, effusive retinitis and even retinal detachment can occur. Noneffusive FIP usually presents with central nervous symptoms depending on which other organs are involved, the sites of the affected nervous system components and the severity of injury. If the lesions spread to the spinal cord, a series of symptoms, such as mild paralysis of the hips, ataxia, hyperesthesia, epilepsy and numbness in the forelimbs, can occur (Pedersen, 2009).

4.2. FECV biotype

Naturally infected kittens develop low fever with intermittent vomiting 3–6 days after infection (Pedersen et al., 1981). At this point, intestinal peristalsis accelerates, and persistent diarrhea occurs for approximately two to five days. During the initial stage, the feces are watery and pasty. As the disease develops, blood can be seen in the feces. Subsequently, sick cats show lethargy and anal swelling. If enteritis is serious, dehydration occurs. However, mortality rates are generally low. Infected cats release virions through their feces for approximately 2–24 months (Pedersen et al., 2008).

Carcasses often show no major pathological changes or injuries, but naturally infected young cats exhibit mesenteric lymph node swelling and intestinal edema, and sometimes exfoliated mucosa cells are observed in the feces of infected cats (Pedersen et al., 1981). Usually, cats gradually recover from intestinal injury. Microscopic examinations have shown that mature epithelial cells at the tips of small intestinal villi form syncytial bodies and that the adjacent villi fuse; in addition, the infected epithelial tissue and villi detach, and columnar epithelium hyperplasia and villus reconstruction subsequently occur. Microscopic examinations have also shown that the neutrophil representation in blood falls below 50%.

5. Pathogenesis of FCoV infection

Envelopment of FECV particles increases their extracellular lability, especially once the virus is shed from the cat, usually in the fecal matter (Evermann et al., 1991). The virus appears to be trypsin- and host protease-dependent, although it has an envelope and pH resistance. These properties maintain FECV infectivity as the virus passes through the cat's intestinal tract (McKeirnan et al., 1987). FECV tends to invade and replicate in columnar epithelial cells from the lower middle intestine to the cecum and then gradually relocate to the mesenteric lymph nodes, tonsils and upper respiratory tract through blood and lymph (Kipar et al., 2010). This mechanism is because FECV can infect monocytes, causing systemic infection, and proliferate in intestinal epithelial cells (Meli et al., 2004; Porter et al., 2014). However, monocyte/macrophage infection may cause the virus to mutate. Researchers have found that viruses that replicate in intestinal tissue cells, monocytes, and macrophages often mutate, making the monocytes and macrophages more susceptible to the virus and thus ensuring that the virus can maintain sustained and effective replication in these cell types (Dewerchin et al., 2005; Pedersen, 2009). Increased replication gradually leads to the production of virulent strains.

Regarding the origin and pathogenesis of FIP, the widely accepted theory involves internal mutation (Dye and Siddell, 2007; Myrrha et al., 2019; Pedersen et al., 2009) (Fig. 2). In this process, FECV mutates into FIPV within the infected host, which causes disease (Pedersen et al., 2014). Although FECV and FIPV are two different biotypes with very different virulence, sequences of these two viruses isolated from cats kept in the same house have been found to be highly similar, while sequences of both viruses isolated from cats kept in different regions and houses have been shown to be mutated (Herrewegh et al., 1995; Pedersen et al., 1981). Approximately 5% of cats with FIPV infection have persistent FECV infection (Lewis et al., 2015). The current view is that FIPV internal mutation arises from the monocyte/macrophage tropism mechanism mentioned above; i.e., FECV usually proliferates in its normal host cell (mature intestinal epithelium), but once it adapts to proliferate in monocytes/macrophages, a change in the virulence of the virions and a change in biotype will occur, resulting in FIPV (Pedersen et al., 2012; Tekes and Thiel, 2016). However, cell tropism transformation does not reveal the true mechanism of internal mutation. Accordingly, researchers have attempted to explain the mechanism of the mutation at the molecular and genetic levels. The FCoV accessory gene 3c and the S gene are associated with internal variation in FIPV (Tekes and Thiel, 2016). Early studies showed that in FECV, the 3c genes are always complete, while in FIPV, more than two-thirds of the 3c genes contain mutations, including deletions and point mutations. These mutations prevent the 3c gene from expressing an intact 3c protein. The prevailing view is that mutations in the 3c gene do not affect FIPV virulence, but it cannot be excluded that such mutations lead to changes in FECV cell tropism that increase adaptability in monocytes and macrophages and lead to the development of FIPV (Pedersen, 2009). Some researchers have deduced that the nonstructural glycoprotein 7b, encoded by ORF7b, determines FCoV virulence (Vennema et al., 1992). However, Myrrha et al. found that the highly conserved 7b gene could not be used as a valid gene to distinguish virulent (FIPV) and avirulent (FECV) biotype switching (Myrrha et al., 2019). The other genes encoding proteins 3a, 3b and 7a also affect the adaptation of viruses to cells (Haijema et al., 2004); however, there is no evidence that specific mutations in these genes are closely related to the development of FIP (Bank-Wolf et al., 2014). Recent studies on FIP pathogenesis have highlighted the importance of S genes. Coronavirus S proteins play a key role in receptor binding for viral invasion, and the transformation from FECV to FIPV involves changes in target cell tropism. Therefore, scientists speculate that changes in FCoV biotypes may also be associated with mutations in S genes. Previous studies have shown that the characteristic differences in S proteins between FECV and FIPV include mutations at two amino acid sites: a methionine (M)-to-leucine (L) mutation at 1058 and a serine (S)-to-alanine (A) mutation at 1060 (Chang et al., 2012). Both sites are located at the beginning of the heptad repeats 1 (HR1) region of the S protein and are related to cellular tropism (Bank-Wolf et al., 2014; Jaimes and Whittaker, 2018; Ouyang et al., 2022; Shirato et al., 2018). Mutations at these sites lead to changes in the cellular tropism of the virus and increase virus adaptability to monocytes and macrophages in over 95% of cases. Therefore, it has been speculated that the two mutation sites might be associated with the differentiation of FIPV from FECV. However, different researchers have different opinions. Porter and coworkers further studied the viral genome and found that an M1058L mutation might not cause FIPV transformation, but it can induce a change in cell tropism. They speculated that the M1058L substitution is an indicator only of the ability of FCoV to spread systemically from the gastrointestinal tract, not an indicator for confirmation of FIP (Porter et al., 2014). Later, Louise Longstaff and coworkers reported that 65% of cats positive for FCoV had leucine at position 1058, while 12% had methionine. At the 1060 site, only 12% had methionine, while 6% had alanine (Longstaff et al., 2017). In 2019, Laura Sangl and coworkers (Sangl et al., 2019) detected FCoV mutations in paraffin-embedded tissues from cats with FIP and control cats. They found that 23 of 34 samples from cats with FIP had the amino acid substitution M1058L in the S protein, while none had the substitution S1060A. Among the 11 other samples from cats with FIP, one sample had a mixed population of nonmutated FECV and FIPV (mixed genotype); 7 samples showed a low viral burden below the limit of detection; one sample showed a high FCoV load (in a cat with FIP), but the amino acid substitution was not found; and two samples were negative for viral RNA. In addition, none of the control animals showed FCoV amino acid substitutions. Therefore, the findings showed that the substitution M1058L in the S protein was present in a majority of tissue samples from cats with FIP, consistent with Chang et al.’s report (Chang et al., 2012). Recently, Decaro et al. (2021) also performed mutation analysis of the S protein in Italian FIPV and FECV sequences. They analyzed a total of 40 FCoV strains, including 19 strains from effusions or tissue samples of FIP cats and 21 strains from fecal samples of non-FIP cats. They found that 16/18 FCoV I and 1/1 FCoV II strains associated with FIP displayed the M1058L mutation, while the S1060A mutation was presented by two FIPV strains, including the single FCoV II strain and one of the two FCoV I strains, which confirms that residue 1058 is a key residue for the FECV/FIPV shift or at least for the systemic spread of the virus. Although there is still debate regarding whether the M1058L and S1060A substitutions in the S protein are responsible for the genotype switch, increasing numbers of studies support the idea that M1058L is more helpful than S1060A for explaining the genotype switch from FECV to FIPV. Lin further identified that approximately 94.68% (89/94) of FCoV isolates had a mutation of methionine to leucine at position 1058 in the spike protein in China between 2019 and 2021 (Lin et al., 2022). In addition to specific point mutations in S proteins, other mutations at the spike protein cleavage site might also be useful for distinguishing FIPV and FECV genetically. Licitra et al. (2013) reported that a furin cleavage site in the region between the receptor-binding (S1) and fusion (S2) domains of the spike of FCoV was closely associated with the distinction of FECV and FIPV. They revealed that all FECVs had a conserved furin cleavage motif of R-R-S/A-R-R-S, while as many as 10 out of 11 FIPVs contained amino acid substitutions at the cleavage site itself or in close proximity to the furin cleavage site. The researchers deduced that mutation of the S1/S2 locus and modulation of a furin recognition site normally present in the S gene of FECVs are critical contributing factors for the development of FIP, ideas that have been supported by other researchers (Tekes and Thiel, 2016). Additionally, it has been reported that S1/S2 cleavage site mutations, including R790S, R793S and R790G, in FCoV might be related to the circulating virulent-avirulent theory of FIP pathogenesis (Healey et al., 2022). Hehao Ouyang et al. further found that the S1/S2 cleavage site in FCoV contains a conserved monobasic sequence (R↓S) preferred by trypsin-like enzymes, and FIPV included a higher minimal R-X-X-R recognition motif of furin (41.94%) than did FECV (9.1%) (Ouyang et al., 2022) (Fig. 1b).

Fig. 2.

Schematic representation of the internal mutation theory.

After FCoV acquires macrophage tropism, it can spread systemically. However, not all systemically spreading FCoV strains are pathogenic. Some mutant viruses may cause abortive infection due to the hosts’ strong and rapid cellular immune response even though the virus can spread to systemic tissues, including the mesenteric lymph nodes, the intestinal serosal surface and sometimes the pleura and omentum (Barker et al., 2017; Pedersen, 2009). A number of viruses can also reach the meninges (especially the posterior ventral surface of the brain), the ependyma surrounding the ventricles, the notochord epidural space abutting the brain, and the uveas and retinas of the eyes (Pedersen, 2009).

FCoV replication is slow in the first two weeks of infection, but at 10–21 days after infection, macrophage penetration and viral invasion and replication begin to accelerate rapidly. Surprisingly, cats can be infected by the different biotypes of FCoV (FIPV and FECV) at the same time (Pedersen, 2009). Another surprising finding is that FCoV has antibody-dependent enhancement (ADE) action (Corapi et al., 1995; Takano et al., 2019); the antibody response not only fails to neutralize the virus and prevent it from spreading but also increases the infectivity of the virus. When antibodies bind to the virus, they block binding to the feline aminopeptidase receptor (fAPN) but also increase the chance that the virus will be consumed by macrophages because of the Fc receptors on the antibodies, increasing the risk of infection. Notably, it is generally accepted that fAPN can act as a receptor for the attachment and entry of FCoV II strains but not FCoV I strains (Gao et al., 2022). Therefore, this enhancement of antibody regulation is specific to feline antibodies and macrophages and is related to viral S and M proteins of FCoV II (Dewerchin et al., 2006; Takano et al., 2017).

The pathogenesis of effusive and noneffusive FIP is of great interest to researchers. It has been reported that the occurrence of effusive or noneffusive form depends on immune status (Paltrinieri et al., 1998). Studies have shown that FIPV infections caused by both inoculation in the laboratory and natural infection have very short effusive FIP periods. Observations of experimentally and naturally infected cats have shown that lesions associated with noneffusive FIP are remnants of lesions associated with effusive FIP, which indicates that during the initial stage of FIPV infection, a short episode of effusive FIP might develop first because of the weak immune response to the virus, while with the improvement of the immune response, effusive FIP should gradually transform to noneffusive FIP (Pedersen, 2009). When the immune system is suppressed, noneffusive FIP can transform into effusive FIP. In the final stage of the disease, the occurrence of noneffusive FIP is presumably due to partial protection by the immune system. Whether effusive or noneffusive FIP occurs may be related to whether the body initiates a cellular or humoral immune response in the early stage of infection. Clinical disease symptoms do not occur if cellular immunity is activated early and if the immune response is sufficient, although viral replication can still be detected. At this time, the infection often progresses to noneffusive FIP. If a humoral immune response occurs and cellular immune initiation fails, the final form of the disease is often effusive or wet FIP. Nevertheless, in many cases, the host mounts an immune response in the middle stage of infection, initiating strong humoral immunity and weak cellular immunity (Pedersen, 2009, 2014). At this time, both virus replication and virus eradication occur. As a result, noneffusive FIP develops, and typical granulomas form. In the centers of the granulomas are small aggregations of macrophages that generally do not contain virus antigens or contain only small amounts of virus antigens surrounded by a broad band of lymphocytes or plasma cells. Effusive FIP can lead to the formation of pyogranulomas covered with virus-loaded macrophages and scattered neutrophils and lymphocytes (Pedersen, 2009). Pyogranulomas extend from the surface of an organ into the underlying parenchyma. The distribution of dry FIP granulomas in organs is related to a partial or weak immune status. At the initial stage of FIPV invasion, the virus reaches the brain and eyes through infected monocytes. However, because of the blood‒brain barrier, immune cells have difficulty reaching the brain. If protective immunity is strong enough, virus particles or virus-infected cells are quickly eradicated; if immunity is weak, immune cells are much more effective in dealing with virus particles and virus-infected cells in the core of the body than they will be in the brain or eyes due to the blood‒brain barrier. Thus, lesions in the brain and eyes last longer than those in the abdomen or other parts of the body, which explains why dry FIP is more likely than wet FIP to occur in the brain and eyes (Fig. 3).

Fig. 3.

Schematic representation of the pathogenesis of effusive and noneffusive FIP (Pedersen, 2009).

6. Diagnosis of FCoV infection

6.1. FIP

6.1.1. Clinical symptoms

In cases of effusive FIP, abdominal swelling can be seen. An obvious fluid wave can be felt upon palpation, but it is necessary to consider the differential diagnosis of pregnancy in cats or other circumstances that cause abdominal swelling. Sometimes uveitis and neurological symptoms occur, but an exact diagnosis cannot be made according to only these symptoms. For noneffusive FIP, although lesions in the eye can occur, they are not characteristic. The diagnosis of FIP requires histological, viral antigen, physiological, biochemical, serological and molecular biological examinations after the death of the sick cat (Giori et al., 2011; Thayer et al., 2022).

6.1.2. Blood biochemical indicators

In FIP-affected cats, blood abnormalities, including anemia (both regenerative and nonregenerative anemia), microcytosis (sometimes associated with anemia), lymphopenia (more common in effusive FIP cases), band neutrophilia (sometimes associated with segmented neutrophilia), and thrombocytopenia, are common. Among these abnormalities, microcytosis and band neutrophilia are very common features in FIP. Cats with FIP have a 50% chance of developing lymphopenia, and most cases appear in cats with effusive FIP (Felten and Hartmann, 2019). Many cats have abnormalities in serum biochemistry, mainly hyperproteinemia, hyperglobulinemia (even when serum total protein does not increase), hypoalbuminemia (very common in effusive FIP cases), and hyperbilirubinemia (common in effusive FIP cases). In addition, some potentially involved organs develop azotemia (often in noneffusive FIP cases), and the activity of liver enzymes increases. Among all these biochemical abnormalities, hyperglobulinemia has the highest frequency, at 89% (Riemer et al., 2016). Changes can occur in different combinations, but it should be noted that these features are neither pathognomonic nor specific for FIP, which means that they can occur in any cat with inflammatory disease. Therefore, they can be used only as a reference for FIP in differential diagnosis (Paltrinieri et al., 2002). Hyperglobulinemia can occur either alone or concurrently with hypoproteinemia. Some studies have shown that the ratio of serum albumin to globulin (A:G ratio) has greater diagnostic value than only the γ-globulin or total protein concentration (Hartmann et al., 2003). If the A:G ratio is greater than 0.6–0.8, FIP can generally be ruled out. Only when the A:G ratio is less than 0.4 can FIP be considered (Jeffery et al., 2012; Norris et al., 2005). However, similar to other hematological or serological characteristics, the A:G ratio should be assessed in combination with disease course, medical history, other laboratory diagnostic parameters and findings from molecular diagnostic methods, if possible. It has been reported that an elevated concentration of γ-globulin, as shown by polyclonal antibody or monoclonal antibody detection during serum protein electrophoresis, indicates FIP in cats (Taylor et al., 2010). However, if a cat has any of the aforementioned clinicopathological abnormalities, an immediate diagnosis cannot be made; other clinical, laboratory and imaging findings need to be considered. The levels of acute-phase proteins, such as alpha-1-acid glycoprotein (AGP), serum amyloid A (SAA) and haptoglobin in serum, increase rapidly in FIPV-infected cats (Hazuchova et al., 2017; Paltrinieri et al., 2007). These parameters are helpful in FIP diagnosis, and a negative AGP test can rule out FIP; however, the parameters do not serve as indicators of FIP on their own, since increased APP concentrations can also occur in other inflammatory conditions, during tumor formation, and even in healthy FECV-infected cats (Felten and Hartmann, 2019).

6.1.3. Detection of effusion

Most cats with FIP develop effusion, especially pleural effusion, ascites or a combination of both; they can even develop pericardial effusion (Riemer et al., 2016). Typical effusion fluid in a cat with FIP is viscous and pale yellow, clear or slightly cloudy with clots or strings. Detection of FIP-related effusion through macroscopic or cytologic methods, including determination of cell counts and detection of biochemical markers, is important and can exclude or confirm differential diagnoses (such as lymphoma detected by cytology or bacterial peritonitis or pleurisy detected by bacterial culture), but many characteristics are not specific to FIP (Giori et al., 2011; Stranieri et al., 2018). Although typical effusion fluids from cats with FIP have high protein concentrations (>35 g/L), low A:G ratios and relatively low total numbers of cells (<5000 nucleated cells/mL), these characteristics can also be found in samples of fluid from effusion caused by other pathogens. Even so, when a cat presents typical symptoms, yields abnormal clinical and laboratory test results (e.g., when a young cat has fever, jaundice, anemia or hyperhemoglobinemia) and presents with effusion, effusion samples should be analyzed by both cytology and bacteriological culture. Even though these characteristics are not specific diagnostic indicators of FIP, other conditions can be ruled out, increasing the likelihood of FIP (Felten and Hartmann, 2019).

The Rivalta test is a low-cost, simple and rapid method to detect exudates (Fischer et al., 2012). This method should be used in the diagnosis of effusive FIP. The test has very good sensitivity (at 91–100%) for excluding FIP, which means that whenever the Rivalta test is negative, other potential causes for effusion are much more likely than FIP (Felten and Hartmann, 2019). In particular, when financial conditions limit testing in a large number of cats, the Rivalta test can be used to exclude FIP; it has a good negative predictive value and does not require automated analysis. However, its specificity has been reported to be only 66–81% (Fischer et al., 2012; Hartmann et al., 2003). Since the Rivalta test can be used to assess secondary increases in the numbers of proteins and inflammatory mediators in fluid, it can be used to initially distinguish exudates from transudates (Berti-Bock et al., 1979), although effusion also occurs in bacterially induced peritonitis, pleurisy, and lymphoma. However, it is easy to make a differential diagnosis based on routine cytology and effusion fluid culture.

Another test based on the protein abundance in effusion fluid from cats with FIP is detection of the change in the total number of nucleated cells (∆TNC) by an automatic hematology analyzer (Sysmex XT-2000iV, Sysmex Europe, Norderstedt, Denmark), which can measure leucocytes through two different channels. The ratio of the numbers of leukocytes in the two channels, ∆TNC, is higher in cats with FIP than in healthy cats. This measurement has good diagnostic accuracy. If 1.7 is established as the cutoff, the diagnostic sensitivity reaches 79–90%, and the specificity reaches 94–100%. If the cutoff ranges from 2.5 to 3.4, the specificity increases to 100% (Giordano et al., 2015; Stranieri et al., 2017).

6.1.4. Serologic detection

FcoV infection causes antibody production independent of the development of FIP. For FCoV infection established under laboratory conditions, serum antibodies can be detected from 7 to 28 days after infection (Felten and Hartmann, 2019; Vogel et al., 2010). Anti-FCoV antibodies were first detected in 1976 (Pedersen, 1976b). Since then, many methods have been established for antibody detection, including indirect fluorescent antibody tests (IFATs) using FCoV or porcine infectious gastroenteritis virus (TGEV) as the antigen, virus neutralization (VN), enzyme-linked immunosorbent assay (ELISA), and rapid immunochromatography (RIM) (Addie et al., 2015). Ideally, tests should be performed on different fluids, such as serum and effusion fluid, and high specificity and sensitivity are needed.

Serum antibody testing is the most common method, but it should be noted that the method cannot distinguish between FECV and FIPV antibodies. Therefore, a high antibody titer detected in the blood does not indicate FIP (Pedersen et al., 1981). There is evidence that FCoV and other coronaviruses, such as TGEV and CCV, cross-react (Barlough and Stoddart, 1988). A large proportion of cats (80% of mixed-breed cats) never develop FIP even though they are positive for anti-FCoV antibodies (Addie et al., 2000; Meli et al., 2004). Cats vaccinated against FCoV also produce antibodies (Haijema et al., 2004). Therefore, the value of antibodies in FIP diagnosis is very limited (Wasmoen et al., 1995). If a cat is infected with FIPV, it should be possible to measure antibody titers. However, when the antibody titer is low or at an intermediate level, its diagnostic value for FIP is almost zero (Hartmann et al., 2003). Although an increasing titer can sometimes indicate FIP progression, increases can also occur during FECV reinfection. Therefore, the antibody titer cannot be used as an indicator to confirm FIP diagnosis (Addie et al., 2000). In addition, a negative antibody test does not rule out FIP (Kennedy et al., 2008) because in the late stage of FIP, especially when fulminant effusion is present, antibody titers decrease dramatically and can drop below the minimum detection limit (Addie and Jarrett, 1992). Approximately 10% of cats with FIP do not have detectable antibodies (Hartmann et al., 2003). There are two potential reasons for this. The first is that immune-mediated vasculitis results in infiltration of blood components, including antibodies, into effusion fluid, preventing detection of antibodies in the blood. The second is that a large number of virions bind to antibodies in the final stage of FIP, making the antibodies difficult to detect (Meli et al., 2013; Soma and Ishii, 2004). Therefore, antibody detection in a cat at a certain time point has no value in the diagnosis of FIP. To increase the specificity of antibody detection for the identification of FIP infection, a method for FCoV 7b protein antibody detection has been established, but this method is based on the incorrect assumption that FIPV has a complete 7b gene, while FECV does not. In fact, cats infected with either FIPV or FECV can produce antibodies against 7b, so this detection method does not produce better diagnostic results than other antibody tests (Kennedy et al., 2008).

Effusion fluid from cats with FIP also contains antibodies, and some researchers have used anti-FCoV antibodies to detect effusion (Hartmann et al., 2003; Soma and Ishii, 2004). The sensitivity and specificity of the detection of antibodies from effusion samples can reach 85% and 86%, respectively, after a cat is confirmed to be infected with FIPV by histopathology (Hartmann et al., 2003). Although these numbers seem more promising than those associated with antibody detection, both serum antibody detection and effusion antibody detection have low diagnostic value and are limited by many factors. Notably, the detection of antibodies in effusion fluid is subject to some limitations. For example, antibody detection in both serum and effusion samples from a clinically suspected FIPV-infected cat is valuable, but the antibody titer in effusion fluid is 23% lower than that in serum. In fact, detecting antibodies in effusion samples is not more useful than detecting antibodies in serum.

In cats with FIP presenting with neurological symptoms, certain amounts of antibodies can be detected in cerebrospinal fluid (CSF), although the sensitivity is low. Some researchers have proposed that an antibody titer in the CSF reaching or exceeding 1:640 and detection of RNA may be significant in the diagnosis of FIP (Felten and Hartmann, 2019; Soma et al., 2018).

6.1.5. Immunohistochemical method

Immunohistochemistry (IHC) is a method used to detect antigens via immunological and biochemical staining of tissues with characteristic histopathological changes caused by FIP. This method has been used to detect lesions in cats with FIP for more than 20 years, and the sensitivity of this detection method is 97–100% (Felten and Hartmann, 2019; Rissi, 2018; Tammer et al., 1995). Therefore, this method is considered to be the gold standard for the diagnosis of FIP, and it can be used to suggest the presence of viruses in tissues with histopathological lesions and to diagnose FIP in atypical cases (Drechsler et al., 2011; Pedersen, 2009). However, it should be noted that not all studies of immunohistochemistry find a high sensitivity (Felten et al., 2019). Attention should be also given to the need to prepare non-FCoV antibodies as controls for detection because cat tissues are adhesive, which causes nonspecific antibody binding.

6.1.6. RT‒PCR

RT‒PCR is a routine laboratory method for the detection of RNA virus sequences. Since Li and Scott (1994) first performed RT‒PCR to detect FCoV in 1994, this technique has often been used to detect virus particles in different materials, such as blood, effusion fluid, CSF, other tissues and feces of infected cats (Addie and Jarrett, 2001; Drechsler et al., 2011). It was once thought that in tissues other than gastrointestinal tract tissues, such as blood, effusion fluid and other bodily fluids, a positive result indicated the presence of a virulent FIPV biotype; however, more recent studies have shown that this theory is incorrect since FCoV RNA in tissues outside the gastrointestinal tract is not necessarily attributable to FIPV (Felten and Hartmann, 2019). It is well known that cats with FIP carry more virions than healthy cats with FECV (Can-Sahna et al., 2007). Therefore, another PCR technique, real-time or quantitative RT‒PCR, can be used to efficiently and quickly determine the viral loads in samples (Meli et al., 2004). A positive RT‒PCR result in combination with a high viral load is very helpful in diagnosing FIP. Researchers can also distinguish between effusive FIP and noneffusive FIP according to exudates and quantitative RT‒PCR results. Dunbar recently established a quantitative RT‒PCR method that can be used to identify whether cats with FIP have noneffusive FIP with a sensitivity of 90% and a specificity of 96.1% (Dunbar et al., 2019). Hellemans (2020) recently reported that a real-time fluorescence quantitative PCR technique combined with an immunoperoxidase monolayer assay (IPMA) can be used to identify effusive FIP with high specificity and sensitivity. In addition, the authors used Sanger sequencing to confirm 5 characteristic variant sites in the S protein in effusive FIP. In addition to the previously reported M1058L and S1060A mutations, three other characteristic mutations, I1106T, D1108Y/E/G and T1112S/N, were identified.

6.1.7. Other diagnostic methods

Other diagnostic methods for FIP include ultrasound examination, X-ray imaging, and immunocytochemical analysis (Felten and Hartmann, 2019; Griffin, 2020; Kitagawa et al., 2007). How to further improve the accuracy, sensitivity and specificity of antemortem detection of FIP in cats is a diagnostic problem that remains to be solved.

6.2. FECV

The diagnostic methods for FECV biotype infection are similar to those for FIPV infection and are summarized below.

A definitive FECV biotype diagnosis must be based on clinical and pathological changes, especially lesions characteristic of enteritis, identified by the following methods:

Histological examination: Pathological tissues from cats with suspected FECV infection should be obtained to prepare frozen sections that will be subjected to fluorescence antibody detection. FECV mainly infiltrates small intestinal tissues and mesenteric lymph nodes; sometimes infiltrates the tonsil and thymus tissues; and almost never infiltrates other organs and tissues, such as the spleen, lungs, kidneys and liver. Tissue pieces can also be directly dyed with hematoxylin and eosin, but this type of staining usually does not clearly show histological lesions in the small intestine and other organs (Pedersen et al., 1984).

Antibody detection: Coated antibodies specific to FECV can be used to detect the virus in feces or tissues by ELISA. In addition, antigen coating can be used to detect antibodies in feline serum. Antibody fluorescence labeling and immunofluorescence techniques can be used to detect antibodies in feline serum. Twelve to 14 days after FECV infection, fluorescent antibody titers are generally within the range of 1:25–1:1024. The antibody titers in cats with noneffusive FIP can exceed l:3200, while the titers in cats with effusive FIP range from 1:100∼1:3200; these values can be used as a basis for clinical diagnosis (Pedersen et al., 1981). Surprisingly, cats with high antibody titers are more likely to shed viruses than cats with low antibody titers (Pedersen et al., 2008). In addition, an infected female cat can produce maternal antibodies to protect kittens until the age of 9–10 weeks (Foley et al., 1997a).

Cell culture: The cell lines routinely used in FECV culture are similar to those used in FIPV culture and include Felis catus whole fetus-4 (FCWF-4), porcine testis, Grandall, fc0009 and feline kidney cell lines (Pedersen et al., 1981). The adaptability of FECV strains varies among cell lines; for example, FECV-79–1683 can grow stably in all of the above cell lines, but FECV-UCD can be cultured and passaged only in cats.

RT‒PCR: Detection of the FECV biotype by RT‒PCR has rarely been reported. In 2008, Pedersen and coworkers (Pedersen et al., 2008) first reported the use of real-time fluorescence quantitative PCR to detect the dissemination of FECV after infection. Recently, McKay and coworkers (McKay et al., 2020) reported the use of RT‒PCR to detect FECV in biological samples and revealed an approximately 46% infection rate for samples of feline feces; they also found that the primary virus serotype was FCoV I.

7. Prevention and treatment

There has been a delay in FIPV vaccine development because of ADE, which causes vaccinated animals to produce antibodies that actually benefit virus infection. There is only one FCoV vaccine available; it requires nasal inoculation and is used mainly in the United States and some European countries. The vaccine contains a temperature-sensitive FCoV II mutant strain, DF2-FIPV (Gerber et al., 1990). However, since FCoV I is more prevalent than FCoV II, the vaccine's protective effect remains controversial (Addie et al., 2009). Takano and coworkers (Takano et al., 2014) recently designed a polypeptide vaccine consisting of T helper epitopes derived from the nuclear protein of FIPV KU-2; this vaccine overcomes the drawbacks of ADE. However, the effectiveness of the vaccine still needs to be improved. Routine precautions against FIP include blockade of its route of transmission. If cats with FIP die in a house, it is recommended that the owner thoroughly disinfects the environment and does not obtain a new cat for at least two months so that all the FIPV particles in the environment die or become inactivated. In addition, to ensure a healthy living environment for cats, the dwelling should be kept clean and disinfected regularly, overcrowding should be avoided (no more than 3 cats per room), and stress should be prevented in cats. In addition, care should be taken when cats from different environments mingle, especially when a new cat is introduced, and any FCoV infections should be identified in existing and new cats. Only through routine cleaning can the source of FIP be eliminated (Foley et al., 1997b).

There are no specific drugs to treat FIP, and symptomatic treatment can only relieve symptoms and prolong life. Tylosin and prednisolone were the first reported drugs to be used to cure cats with FIP. However, tylosin has no effect on FIP after a long decade of treatment (Colgrove and Parker, 1971; Pedersen, 2009). It has been reported that some cats have gone into remission after the use of prednisolone and phenylalanine mustard or cyclophosphamide with anti-inflammatory and anti-allergic effects, and injections of antibiotics such as ampicillin and ceftiofur can prevent secondary bacterial infection (Pedersen, 1976a). FIPV is reportedly sensitive to human α- and β-interferon in vitro (Weiss and Toivio-Kinnucan, 1988). In addition, ω-interferon also inhibits FIPV in vitro, and the drug has been used in some countries (Mochizuki et al., 1994). However, a large double-blind experiment indicated that the drug is generally ineffective (Ritz et al., 2007). Takano et al. reported that hydroxychloroquine in association with interferon-omega (IFN-omega) is effective for FIPV treatment in vitro; however, further research is needed to determine whether the drugs can be used clinically (Takano et al., 2020). Other reported drugs include pentoxifylline (an immunomodulator), nelfinavir (an antiviral drug), polyisopentene (an immunostimulant), and ozagrel hydrochloride (a thromboxane synthase inhibitor). The efficacy and safety of these agents need to be evaluated in future clinical trials. Small interfering RNA molecules and cyclosporine A have been shown to inhibit FIPV replication in vitro, but their effects in vivo have not been confirmed (Tanaka et al., 2016). In recent years, new drugs to treat FIP have also emerged. Pederson and coworkers (Pedersen et al., 2018) recently confirmed the efficacy of a 3C-like protease inhibitor (GC376) in experimentally infected cats. However, the data showed that relapse could occur. Zhou reported two additional 3C-like protease inhibitors against FIPV and swine coronavirus, 3-(aminocarbonyl)-1-phenylpyridinium and 2,3-dichloronaphthoquinone, which exhibited significant antiviral capacity against these viruses on Vero cells and feline kidney cells, but their curative effects on cats with FIP were not reported (Zhou et al., 2021a). Pederson and coworkers (Pedersen et al., 2019) also discovered a nucleoside analog (GS-441,524) that can directly interfere with the replication of viral coding genes. As a result, the treatment can be safely and effectively administered to cats naturally or experimentally infected by FIPV (Meli et al., 2022). Recently, a nonnucleoside inhibitor targeting viral RNA polymerase, ERDRP-0519 (ERDRP), was reported to significantly inhibit the replication of FIPV in a dose-dependent manner in vitro (Camero et al., 2022). However, its effect on FIP in vivo is still under investigation.

There are few reports on the prevention and treatment of infection by the FECV biotype, and there is generally no need to administer systemic therapy or antibiotics to treat the disease. If enteritis is serious and a cat becomes dehydrated, the cat should receive fluids for rehydration and nutrition in a timely manner. Beginning at 5 weeks of age, kittens are not infected because of maternal antibody protection. Newly weaned cats that are infected are often asymptomatic but can still spread the disease, causing severe enteritis. Cats of all ages should be separated (Drechsler et al., 2011). Weaned cats should receive appropriate care and treatment only if the severity of the disease can be greatly reduced. Recently, Addie and colleagues reported a new oral medicine named Mutian® Xraphconn (Mutian X), a product marketed to treat cats with FIP that is also being used to successfully stop virus shedding for the first time (Addie et al., 2020b).

8. Conclusions and perspectives

There is still no effective FCoV vaccine, which continues to limit the prevention and treatment of the disease caused by this virus. In addition, the immune mechanism against the virus is still mostly unknown, especially regarding precise cellular epitopes. Improving the accuracy, sensitivity and specificity of antemortem FIP diagnosis in cats is also a problem remaining to be solved. However, research on the pathogenesis, diagnosis, prevention and treatment of FCoV has been helpful in the treatment of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. For example, the 3C-like protease inhibitor GC376 might also be useful for treating patients with coronavirus disease 2019 (COVID-19) because SARS-CoV-2 also has a 3C-like protease (Lu et al., 2022; Vuong et al., 2020). In addition, the active form of remdesivir, GS-441,524, has recently been used to treat SARS-CoV-2 (Paltrinieri et al., 2021). In the future, with the advancement of technology and increasing attention on coronaviruses, there will likely be breakthroughs in research on elaborate epitope vaccines, immune mechanisms and methods for accurate antemortem diagnosis and treatment.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

• Data will be made available on request.