Immunoreactivity for CD34, Desmin, Keratins, KIT, Alpha‐Smooth Muscle Actin, S100, and Vimentin in Malignant Mesenchymal Neoplasms in Guinea Pigs: A Series of 62 Cases From a Single Institution

ABSTRACT

Background

There has been a lack of research regarding diagnostic immunohistochemistry in soft tissue pathology in rodents.

Objectives

We aimed to analyse the immunohistochemical expression of CD34, keratin AE1/AE3, keratin 8/18, desmin, KIT, SMA, S100, and vimentin in various malignant mesenchymal tumours in guinea pigs.

Methods

A total of 62 tumours included 10 fibrosarcomas, 8 myxosarcomas, 8 undifferentiated pleomorphic sarcomas, 6 undifferentiated round cell sarcomas, 6 dedifferentiated liposarcomas (involving deep soft tissue, skin or subcutis), 12 uterine leiomyosarcomas, 4 hemangiosarcomas (involving skin and spleen), and 8 gastrointestinal stromal tumours (GIST).

Results

Desmin and SMA positivity was present in tumours of various lineages, including hemangiosarcoma. CD34 positivity was restricted to hemangiosarcomas, GISTs and one case of undifferentiated pleomorphic sarcoma, whereas KIT positivity was detected only in GISTs. Lesser amounts of keratins positivity were observed in fibrosarcoma, myxosarcoma, undifferentiated pleomorphic sarcoma, and dedifferentiated liposarcoma. S100 positivity was identified in myxosarcoma, undifferentiated pleomorphic sarcoma, and dediffentiated liposarcoma and vimentin positivity was detected in all tumour types.

Conclusions

The strong and diffuse staining pattern of desmin and SMA can be very helpful in distinguishing leiomyosarcoma from its spindle cell mimics. A novel finding is that SMA positivity was identified in malignant endothelium. Immunolabelling for CD34 and KIT provides reliable markers for vascular neoplasms and GISTs. Vimentin does not allow to distinguish between different mesenchymal malignancies in guinea pigs. Occasional positivity for keratins and S100 points to potential pitfalls and emphasised the need for a panel of immunohistochemical stains in the investigation of mesenchymal tumours.

Immunohistochemical expression of CD34, desmin, keratin AE1/AE3, keratin 8/18, KIT, SMA, S100, and vimentin in normal guinea pig skin serving as both positive and negative external tissue control. All images immunohistochemical staining, 200× magnification, scale bar 100 µm. (A) CD34 stained only vascular endothelium (arrow). (B) Desmin stained only muscle cells (arrow). (C) Keratin AE1/AE3 stained only epidermal and adnexal epithelium (arrow). (D) Keratin 8/18 stained only lower parts of the follicles and sebocytes (arrow). (E) KIT stained only mastocytes (arrow). (F) Vimentin stained both mesenchymal and immune cells (arrow). (G) SMA stained only myopericytes and muscle cells (arrow). (H) S100 stained only adipocytes (arrow). NA of the objective lens: 0.50 for 200× magnification.

1. Introduction

In the vast majority of soft tissue sarcomas, if not all, a definitive diagnosis cannot be achieved without the application of ancillary studies such as immunohistochemistry. Endothelial marker CD34, muscle markers desmin and alpha‐smooth muscle actin (SMA), marker of gastrointestinal stromal tumour (GIST), mast cells and myeloid stem cells KIT (CD117), marker of nerve sheath tumours and melanoma S100, and mesenchymal marker vimentin are the most commonly used immunohistochemical antibodies in soft tissue pathology. Unfortunately, none of these markers is entirely specific (Perez‐Montiel et al. 2006; Hasegawa et al. 2000; Miettinen et al. 2006). In addition, positivity for various keratins has been reported in several human mesenchymal malignancies, including leiomyosarcoma of bone, epitheliod sarcoma, and retroperitoneal schwannoma (Romeo et al. 2012; Fanburg‐Smith et al. 2006; Miettinen et al. 2013). During the diagnostic process, reporting immunohistochemical staining results only as ‘positive’ or ‘negative’ is not adequate. Instead, several parameters should be evaluated for each individual immunohistochemical test, namely the intensity and extent of staining and the cellular and tissue localisation of staining. This method of assessment determines the so‐called immunohistochemical staining pattern, which, in addition to morphology, is the key to establishing a correct diagnosis (Goldsmith et al. 2024).

To clarify the immunohistochemical background of our study, we briefly summarise the distribution of analysed markers in normal and neoplastic tissues. Among normal tissues, CD34 has been detected in both vascular and lymphatic endothelial cells, hematopoietic progenitor cells, mast cells, fibroblasts/fibrocytes, interstitial cells of Cajal, endometrial stromal cells, and osteoblasts (Pusztaszeri et al. 2006; Hassanein et al. 2009; Hirose et al. 2003). Desmin has been found in smooth muscle cells, myoblasts, myofibroblasts, reactive mesothelial cells, hepatic stellate cells, and decidualised stromal cells (Capetanaki et al. 1997; Hasteh et al. 2010; Heatley et al. 1996; Nitou et al. 2000). Keratin AE1/AE3 has been observed in epithelial cells, trophoblasts, hepatocytes, bile ducts, and extrafollicular reticular cells of the lymph node (Goddard et al. 1991; Daya and Sabet 1991; Olson et al. 2016; van Eyken et al. 1987). Keratin 8/18 has been proven in non‐stratified epithelial cells, trofoblasts, lymphatics and venules (Miettinen and Fetsch 2000; Austgulen et al. 2002). KIT has been detected in interstitial cells of Cajal, melanocytes, breast epithelium, hematopoietic progenitor cells, and germ cells (Miettinen and Lasota 2005). SMA has been found in smooth muscle cells, myofibroblasts, sweat glands, pericytes, osteoblasts, hepatic stellate cells, decidual stromal cells, chondrocytes, breast myoepithelial cells, and salivary glands (Zhang et al. 2003; Kana et al. 2006; Draeger et al. 1991; Kinner and Spector 2002). S100 has been observed in melanocytes, glial cells, neurons, Schwann cells, adipocytes, chondrocytes, myoepithelial cells, dendritic cells, interdigitating dendritic cells, and Langerhans cells (Huber 1997). Vimentin has been found in mesenchymal cells, blood and immune cells, melanocytes, endometrial glandular cells, and Leydig cells and Sertoli cells (Battifora 1991; Ramos et al. 2020). Regarding neoplastic tissues, CD34 positivity has been found, among others, in hemangioma, angiosarcoma, hemangiopericytoma, dermatofibrosarcoma protuberans, epithelioid sarcoma, solitary fibrous tumour, and alveolar soft part sarcoma (Meis‐Kindblom and Kindblom 1998; Fina et al. 1990; Aiba et al. 1992). Desmin positivity has been detected, among others, in skeletal and smooth muscle tumours, skeletal and smooth muscle components of other tumours, fibroblastic/myofibroblastic tumours, and fibrohistiocytic tumours (Truong et al. 1990; Bian et al. 2019). Keratin AE1/AE3 positivity has been observed, among others, in most carcinomas, undifferentiated pleomorphic sarcoma, epithelioid sarcoma, leiomyosarcoma of bone, synovial sarcoma, chondroblastoma, chordoma, and mesothelioma (Kandukuri et al. 2017; Romeo et al. 2012). Keratin 8/18 positivity has been demonstrated in various adenocarcinomas, hepatocellular carcinoma, epithelioid sarcoma and synovial sarcoma (Cheng et al. 2022; Saito et al. 2002). KIT positivity has been detected, among others, in gastrointestinal stromal tumour, epithelioid sarcoma, angiosarcoma, leiomyosarcoma, melanoma, rhabdomyosarcoma, osteosarcoma and synovial sarcoma (Miettinen and Lasota 2005; Garcia et al. 2006). SMA positivity has been observed, among others, in leiomyoma, leiomyosarcoma, rhabdomyosarcoma, synovial sarcoma, myofibroblastic sarcoma, desmoplastic melanoma and undifferentiated pleomorphic sarcoma (Abeler and Nenodovic 2011; Mentzel and Kuhnen 2006). S100 positivity has been found, among others, in nerve sheath tumours, malignant peripheral nerve sheath tumours, adipocytic, chondroid and notochordal tumours, synovial sarcoma, melanoma, myoepithelial tumours, and salivary gland tumours with myoepithelial differentiation (Yamaguchi et al. 2003; Nagao et al. 2012). Vimentin positivity has been demonstrated, among others, in mesenchymal tumours of soft tissues or other organs, mesothelioma, melanoma, renal cell carcinoma, and uterine carcinosarcoma and endometrial carcinoma (Lin and Liu 2014; Kuroda et al. 2003).

Our study deals with soft tissue neoplasms, the incidence of which in guinea pigs ranges from 11 to 41%. According to available data, the most common soft tissue tumour is subcutaneous lipoma, with incidence of up to 16% reported. This is followed by cutaneous and subcutaneous soft tissue sarcoma of myofibroblastic origin, such as fibrosarcoma and myxosarcoma, with an incidence of around 10%. In contrast, ossifying fibroma, fibrolipoma, schwannoma, liposarcoma, hemangiosarcoma, undifferentiated sarcoma, GIST or leiomyosarcoma occur with a lower incidence, not exceeding 5% (Nešpor et al. 2023; Jelínek 2003; Bertram et al. 2025; Blumenthal and Rogers 1965; Kitchen et al. 1975). Although prognostic data for mesenchymal malignancies in guinea pigs are limited, one autopsy study reported a lower frequency of metastasis for several malignancies that behaved aggressively in other animal species, including splenic hemangiosarcoma (Bertram et al. 2025).

2. Materials and Methods

2.1. Tissue Specimens and Study Group

Our study analysed a series of 62 tissue samples obtained from different types of malignant mesenchymal tumours in guinea pigs. Tissue samples, pathology reports and slides were retrospectively retrieved from our institution in the study period 2016–2025. All cases were searched electronically from the institutional registry from 2016 to 2025 using the following keywords: ‘guinea pig’, ‘soft tissues’, ‘soft tissue sarcoma’, ‘sarcoma’, ‘malignant mesenchymal tumour’, ‘mesenchymal malignancy’, ‘fibrosarcoma’, ‘myxofibrosarcoma’, ‘undifferentiated sarcoma’, ‘liposarcoma’, ‘leiomyosarcoma’, ‘angiosarcoma’, ‘gastrointestinal stromal tumour’. Using these keywords, we found a total of 65 cases of malignant mesenchymal tumours, of which 62 cases were included and 3 cases were excluded from the study. The reason for exclusion was the uncertain histogenesis and malignant potential of these 3 tumours. All animals underwent surgery at different clinical workplaces.

Tissue fixation and further histoprocessing were performed according to standard procedures. In brief, immediately after excision, tissue samples were placed in a container filled with 10% neutral buffered formalin for 36 h. Fixation was followed by dehydration, using increasing concentrations of ethanol, clearing of the tissue with xylene, and embedding in paraffin wax. Subsequently, 4–5 µm‐thick sections were cut from each paraffin block and stained with hematoxylin and eosin. Diagnosis was established independently by 3 histopathologists. Interobserver disagreements were resolved by consensus. Finally, representative sections of each tumour sample were selected for immunohistochemical analysis.

Classification was performed according to the latest edition of the World Health Organization (WHO) classification of soft tissue and bone tumours (WHO classification of tumours editorial board 2020) and according to the Histological Classification of Mesenchymal Tumours of Skin and Soft Tissues of Domestic Animals (Hendrick et al. 1998). Ancillary immunohistochemistry with CD34, desmin, keratin AE1/AE3, keratin 8/18, KIT, SMA, S100, and vimentin was used to establish the final diagnosis. Immunohistochemical staining was essential during the diagnostic process; none of the tumours were diagnosed based on morphology alone. In addition to the morphology, CD34 positivity was essential for the diagnosis of hemangiosarcoma, SMA and desmin positivity for the diagnosis of leiomyosarcoma, KIT positivity for the diagnosis of gastrointestinal stromal tumour, KIT, CD34, and desmin negativity for the diagnosis of fibrosarcoma, S100 negativity for the diagnosis of myxosarcoma, and possible focal, not diffuse, positivity of various markers for the diagnosis of undifferentiated sarcoma. Importantly, for all tumour tissue samples analysed, the anatomical location represented the primary site of origin. Metastases were not included in our study.

2.2. Immunohistochemistry

For immunohistochemistry, specially coated microscopic slides for adhesion (Dako, code number K8020) were used. The following antibodies were employed: CD34, desmin, keratin AE1/AE3, keratin 8/18, KIT, SMA, S100, and vimentin (detailed information is provided in Table 1). To eliminate nonspecific staining, Protein Bloc Serum‐Free (Agilent Dako, code X0909, USA) was first applied for 5 min. Prior to immunohistochemical staining, heat‐induced antigen retrieval was performed in EnVision FLEX Target Retrieval Solution (Agilent Dako, code number K8004, USA) (pH 9) for 20 min at 97°C using the PT link pretreatment module (Agilent Dako, USA). The incubation time with anti‐keratin AE1/AE3, anti‐keratin 8/18, anti‐CD34, anti‐desmin, anti‐SMA, anti‐S100, and anti‐vimentin antibodies was set to 20 min, and with anti‐KIT antibody to 60 min. The primary antibodies were visualised using the Dako EnVision FLEX detection system (EnVision FLEX /HRP, code number SM802, Agilent Dako, USA, and EnVision FLEX DAB+ Chromogen, code number DM827, Agilent Dako, USA).

TABLE 1.

Summary of antibodies used for immunohistochemistry.

Antibody	Clone (animal)	Dilution	Manufacturer	Code number
CD34	QBEnd 10 (mouse)	Predilute	Dako (Denmark)	GA632
Desmin	D33 (mouse)	Predilute	Dako (Denmark)	IR606
Keratin AE1/AE3	AE1/AE3 (mouse)	Predilute	Dako (USA)	IR053
Keratin 8/18	EP17/EP30 (rabbit)	Predilute	Dako (USA)	IR094
KIT (CD117)	EP10 (rabbit)	Predilute	Bio SB (USA)	BSB‐9061‐CS
SMA	1A4 (mouse)	Predilute	Dako (Denmark)	1R611
S100	Polyclonal (rabbit)	Predilute	Dako (Denmark)	IR504
Vimentin	V9 (mouse)	Predilute	Dako (Denmark)	IR630

Abbreviation: SMA, smooth muscle actin.

External positive tissue controls were performed for all immunohistochemical assays, guinea pig skin was used for this purpose. The following tissue elements were selectively stained with individual antibodies: CD34 stained endothelial cells (Figure 1A), desmin stained smooth muscle cells (Figure 1B), keratin AE1/AE3 stained epidermal and adnexal epithelium (Figure 1C), keratin 8/18 stained the lower parts of the follicles and sebocytes (Figure 1D), KIT stained mast cells (Figure 1E), vimentin stained mesenchymal cells and blood/immune cells (Figure 1F), SMA stained myopericytes and smooth muscle cells (Figure 1G), and S100 stained adipocytes and nerve fibres (Figure 1H). Negative controls were prepared by incubating samples with antibody diluent replacing the primary antibody (Dabbs 2019), resulting in clear/unstained slides. We thus verified that non‐specific/false‐positive background staining was not present. Validation of the antibodies for use in guinea pigs was performed not only by providing guinea pig external positive and negative tissue controls, but also by comparing guinea pig external tissue controls with validated human external tissue controls, which yielded identical staining results for both guinea pig and human tissues (skin was used as the control tissue in both animal species). Immunohistochemically stained slides were evaluated using a BX53 microscope and a Promican HDMI 16 camera (Olympus, Tokyo, Japan).

FIGURE 1.

Immunohistochemical expression of CD34, desmin, keratin AE1/AE3, keratin 8/18, KIT, SMA, S100, and vimentin in normal guinea pig skin serving as both positive and negative external tissue control. All images immunohistochemical staining, 200× magnification, scale bar 100 µm. (A) CD34 stained only vascular endothelium (arrow). (B) Desmin stained only muscle cells (arrow). (C) Keratin AE1/AE3 stained only epidermal and adnexal epithelium (arrow). (D) Keratin 8/18 stained only lower parts of the follicles and sebocytes (arrow). (E) KIT stained only mastocytes (arrow). (F) Vimentin stained both mesenchymal and immune cells (arrow). (G) SMA stained only myopericytes and muscle cells (arrow). (H) S100 stained only adipocytes (arrow). NA of the objective lens: 0.50 for 200× magnification.

2.3. Evaluation of Immunostaining

For CD34, only membranous staining was considered positive (Natkunam et al. 2000). For desmin, keratin AE1/AE3, keratin 8/18, KIT and vimentin, only cytoplasmic staining was considered positive (Miettinen and Lasota 2005; Wick 1988; Miettinen 1993; Azumi and Battifora 1987). For SMA, both cytoplasmic and membranous staining was considered positive (Jones et al. 1990). For S100, both cytoplasmic and nuclear staining was considered positive (Donato 1999). Two parameters were evaluated for all tumour samples—the percentage of positive cells and the intensity of staining. Evaluation was performed independently by three histopatologists using a light microscope at 200× magnification. Any discrepancies were resolved by consensus.

To analyse marker expression in the entire tumour sample, a large number of areas were studied and the percentage of marker‐positive cells in the analysed areas was recorded. More than 250 tumour cells were examined in each area at 200× magnification. While at least 15 areas were studied in smaller whole‐mount sections, up to forty areas were studied in larger whole‐mount sections at 200× magnification. The percentage of marker positive cells was reported as a cumulative value. Necrotic tissue was not included in the immunohistochemical analysis. The percentage of marker positive cells was categorised into five levels: 0/negative (0–4%), 1+ (5–25%), 2+ (26–50%), 3+ (51–75%) and 4+ (76–100%). A similar method of assessment has been used in previous studies (Lenz et al. 2011; Lenz et al. 2023). The intensity of staining was distinguished between weak, barely discernible at 40× magnification, moderate, clearly discernible at 40× magnification, and strong, very clearly discernible at 40× magnification. If heterogeneous staining intensity was detected in the tumour, the resulting score corresponded to the average of marker positive cells in the analysed area. Importantly, marker positivity was defined as expression in more than 5% tumour cells with moderate or strong staining intensity. The same criteria for marker positivity were used in some previous studies (Lenz et al. 2023; Yoshida et al. 2018). Within each category of staining intensity which was considered positive, slight cell‐to‐cell heterogeneity was observed, ruling out a false‐positive immunoreaction.

3. Results

3.1. Histopathological Features of Malignant Mesenchymal Tumours (Table 2)

TABLE 2.

Summary of tumour types, sites of involvement and number of cases.

Tumour type	Site	N
Fibrosarcoma	DST trunk/extremity	10
Myxosarcoma	Skin/subcutis, DST trunk/extremity	6
Undifferentiated pleomorphic sarcoma	DST extremity	8
Undifferentiated round cell sarcoma	DST extremity	6
Dedifferentiated liposarcoma	Subcutis	6
Leiomyosarcoma	Uterus	12
Hemangiosarcoma	Skin, spleen	4
GIST	Caecum *	8

Abbreviations: DST, deep soft tissue; GIST, gastrointestinal stromal tumour; N, number of cases.

^*Three cases of GIST were located in the caecum, while in the remaining five cases the primary origin could not be verified.

Our series of 62 malignant mesenchymal tumour samples included 10 cases of fibrosarcoma, 8 cases of myxosarcoma, 8 cases of undifferentiated pleomorphic sarcoma, 6 cases of undifferentiated round cell sarcoma, 6 cases of dedifferentiated liposarcoma, 12 cases of leiomyosarcoma, 4 cases of hemangiosarcoma, and 8 cases of GIST. All cases of fibrosarcoma, undifferentiated pleomorphic sarcoma, and undifferentiated round cell sarcoma, and 2 cases of myxosarcoma developed in deep soft tissue of the extremities or trunk. Two cases of myxosarcoma arose in the subcutaneous tissue while the other cases affected the skin. All dedifferentiated liposarcomas were located in the subcutis. All cases of leiomyosarcoma were of uterine origin. Two cases of hemangiosarcoma affected the skin and two cases developed in the spleen. Three cases of GIST were located in the caecum, while in the remaining five cases the primary origin could not be verified. Descriptive characteristics of individual tumours are summarised in the supplemental table.

At the microscopic level, fibrosarcoma was a monomorphic medium to highly cellular spindle cell tumour that showed no more than moderate nuclear atypia and formed intersecting fascicles in a herringbone or storiform fashion (Figure 1A). Undifferentiated pleomorphic sarcoma was composed of highly atypical tumour cells, including bizarre multinucleated giant cells, with patternless architecture and was devoid of myxoid areas (Figure 1B). Undifferentiated round cell sarcoma consisted of sheets of relatively uniform round to ovoid cells with a high nuclear‐to‐cytoplasmic ratio (Figure 1C). All subcutaneous liposarcomas were qualified as dedifferentiated liposarcoma as they had a component of well‐differentiated liposarcoma that abruptly transitioned into an undifferentiated non‐lipogenic component (Figure 1D). Myxosarcoma was characterised by variable nuclear pleomorphism, intercellular myxoid stroma, and obvious elongated, curvilinear, thin‐walled vessels with a perivascular tumour growth (Figure 1E). Two cases of myxosarcoma closely resembled human low‐grade fibromyxoid sarcoma, as they were characterised by a poorly developed vascular network with perivascular sclerosis, collagenised areas, and myxoid nodules of bland spindle cells that occasionally formed abortive whorls (Figure 1F). All cases of uterine leiomyosarcoma were well or moderately differentiated spindle cell tumours with fascicular growth (Figure 2A). Cutaneous hemangiosarcoma consisted of well‐formed anastomosing vessels lined by atypical endothelial cells (Figure 2B), whereas splenic hemangiosarcoma was composed of both vasoformative and solid areas with focal necrosis (Figure 2C). All GISTs were spindle cell tumours. While 6 cases were characterised by parallel rows of tumour cell nuclei and perinuclear cytoplasmic vacuolisation, which closely resembled the human palisaded‐vacuolated subtype of GIST (Figure 2D, E), two cases consisted of densely arranged, uniform spindle cells without significant nuclear atypia, which corresponded to the human hypercellular subtype of GIST (Figure 2F). In addition, a multinodular/plexiform growth pattern was observed in 2 cases of the palisaded‐vacuolated subtype of GIST. The following histological features were useful for revealing the malignant nature of the tumours: poor demarcation, infiltrative growth, increased mitotic activity, atypical mitoses, coagulative necrosis, moderate to severe nuclear atypia with hyperchromasia, lymphovascular invasion, and perineural involvement (Miettinen 1993) (Figure 3).

FIGURE 2.

Morphological features of malignant mesenchymal tumours in guinea pigs. All images hematoxylin‐eosin staining. (A) Fibrosarcoma, deep soft tissue of the trunk. The tumour is composed of cellular spindle cell fascicles arranged in a herringbone pattern (200× magnification, scale bar 100 µm). (B) Undifferentiated pleomorphic sarcoma, deep soft tissue of the extremity. The mass contains highly atypical tumour cells with patternless architecture (400× magnification, scale bar 50 µm). (C) Undifferentiated round cell sarcoma, deep soft tissue of the extremity. The neoplastic cells are monomorphous with round to ovoid nuclei and grow in cellular sheets (400× magnification, scale bar 50 µm). (D) Dedifferentiated liposarcoma, subcutis. Well differentiated liposarcoma (arrow) shows an abrupt transition to non‐lipogenic sarcoma (asterisk) (400× magnification, scale bar 50 µm). (E) Myxosarcoma, skin. Spindle to stellate cells are dispersed in a myxoid background containing elongated curvilinear thin‐walled vessels (arrow) (200× magnification, scale bar 100 µm). (F) Myxosarcoma resembling human low‐grade fibromyxoid sarcoma, deep soft tissue of the extremity. The mass consists of bland spindle cells growing in whorling pattern and forming collagenised areas that alternate with myxoid nodules. Arteriole‐sized vessels with perivascular sclerosis are also present (arrow) (200× magnification, scale bar 100 µm). NA of the objective lens: 0.50 for 200× magnification and 0.75 for 400× magnification.

FIGURE 3.

Morphological features of malignant mesenchymal tumours in guinea pigs. All images hematoxylin‐eosin staining. (A) Leiomyosarcoma, uterus. Spindle‐shaped cells with mild to moderate atypia are set in long intersecting fascicles (200× magnification, scale bar 100 µm). (B) Hemangiosarcoma, skin. Well‐formed anastomosing vessels are lined by atypical endothelial cells (200× magnification, scale bar 100 µm). (C) Hemangiosarcoma, spleen. Neoplastic cells form ill‐defined vascular channels and solid sheets and necrotise focally (arrow) (400× magnification, scale bar 50 µm). (D) Gastrointestinal stromal tumour, palisaded‐vacuolated subtype, caecum. The mass consists of spindle cells with nuclei arranged in parallel rows (arrow); perinuclear cytoplasmic vacuolisation is also evident (200× magnification, scale bar 100 µm). (E) Gastrointestinal stromal tumour, palisaded‐vacuolated subtype, caecum. Vague fascicles of spindle cells with apparent perinuclear cytoplasmic vacuolisation (arrow) (400× magnification, scale bar 50 µm). (F) Gastrointestinal stromal tumour, hypercellular subtype, caecum. Intersecting fascicles consist of densely arranged, uniform spindle cells without significant nuclear atypia (200× magnification, scale bar 100 µm). NA of the objective lens: 0.50 for 200× magnification and 0.75 for 400× magnification.

3.2. Analysis of CD34 in Malignant Mesenchymal Tumours (Table 3)

TABLE 3.

Summary of immunohistochemical expression of CD34, desmin, keratin AE1/AE3, and keratin 8/18 in malignant mesenchymal tumours analysed.

Tumour type	CD34 *	Desmin *	Keratin AE1/AE3 *	Keratin 8/18 *
Fibrosarcoma	0/10 (0%)	0/10 (0%)	0/10 (0%)	7/10 (70%)
Myxosarcoma	0/8 (0%)	3/8 (38%)	2/6 (25%)	3/8 (38%)
Undifferentiated pleomorphic sarcoma	1/8 (13%)	8/8 (100%)	4/8 (50%)	5/8 (63%)
Undifferentiated round cell sarcoma	0/6 (0%)	3/6 (50%)	0/6 (0%)	0/6 (0%)
Dedifferentiated liposarcoma	0/6 (0%)	3/6 (50%)	0/6 (0%)	2/6 (33%)
Leiomyosarcoma	0/12 (0%)	12/12 (0%)	0/12 (0%)	0/12 (0%)
Hemangiosarcoma	4/4 (100%)	2/4 (50%)	0/4 (0%)	0/4 (0%)
GIST	6/8 (75%)	2/8 (25%)	0/8 (0%)	0/8 (0%)

Abbreviation: GIST, gastrointestinal stromal tumour.

^*Positive cases.

Our study confirmed the usefulness of CD34 in the diagnosis of vascular malignancies and GISTs. Positivity was detected in 11/62 tumour samples (18%). Of all the different types of malignancies analysed, CD34 positivity was found only in 1/8 undifferentiated pleomorphic sarcoma (13%), 4/4 hemangiosarcomas (100%) (Figure 4A), and 6/8 GISTs (75%) (Figure 4B). The extent/intensity of staining was classified as 1+/moderate in 1/1 positive undifferentiated pleomorphic sarcoma, and as 4+/strong in 4/4 hemangiosarcomas (100%) and 6/6 positive GISTs (100%).

FIGURE 4.

Immunohistochemical expression of CD34, desmin and keratin AE1/AE3 in malignant mesenchymal tumours in guinea pigs. All images immunohistochemical staining. (A) Hemangiosarcoma, skin. Diffuse and strong immunoreactivity for CD34 (200× magnification, scale bar 100 µm). (B) Gastrointestinal stromal tumour, caecum. Diffuse and strong immunoreactivity for CD34 (100× magnification, scale bar 200 µm). (C) Leiomyosarcoma, uterus. Diffuse and strong immunoreactivity for desmin (400× magnification, scale bar 50 µm). (D) Hemangiosarcoma, spleen. Diffuse and strong immunoreactivity for desmin (200× magnification, scale bar 100 µm). (E) Myxosarcoma, subcutis. Multifocal and moderate immunoreactivity for keratin AE1/AE3 (400× magnification, scale bar 50 µm). (F) Undifferentiated pleomorphic sarcoma, deep soft tissue of the extremity. Multifocal and moderate immunoreactivity for keratin AE1/AE3 (200× magnification, scale bar 100 µm). NA of the objective lens: 0.30, 0.50 and 0.75 for 100×, 200× and 400× magnification, respectively.

3.3. Analysis of Desmin in Malignant Mesenchymal Tumours (Table 3)

Desmin was found in various malignancies, with its highest amount detected in smooth muscle tumours, proving to be diagnostically useful finding. Positivity was detected in 33/62 tumour samples (53%). Desmin positivity was detected in 0/10 fibrosarcomas (0%), 3/8 myxosarcomas (38%), 8/8 undifferentiated pleomorphic sarcomas (100%), 3/6 undifferentiated round cell sarcomas (100%), 3/6 dedifferentiated liposarcomas (50%), 12/12 leiomyosarcomas (Figure 4C), 0/2 cutaneous hemangiosarcomas (0%), 2/2 splenic hemangiosarcomas (100%) (Figure 4D), and 2/8 GISTs (25%). The extent of staining was classified as 1+ in 3/8 undifferentiated pleomorphic sarcomas (38%), 2/3 positive dedifferentiated liposarcomas (67%), and 2/2 positive GISTs (100%), as 2+ in 3/3 positive myxosarcomas (100%), 1/8 undifferentiated pleomorphic sarcomas, 3/3 positive undifferentiated round cell sarcomas (100%), and 1/3 positive dedifferentiated liposarcoma (33%), and as 4+ in 4/8 undifferentiated pleomorphic sarcomas (50%), 12/12 leiomyosarcomas (100%), and 2/2 positive splenic hemangiosarcomas (100%). Staining intensity was strong in 12/12 leiomyosarcomas (100%) and 2/2 positive hemangiosarcomas, strong to moderate in 3/3 positive myxosarcomas (100%), 8/8 undifferentiated pleomorphic sarcomas (100%), and 3/3 positive undifferentiated round cell sarcomas (100%), moderate in 3/3 positive dedifferentiated liposarcomas (100%), and moderate to weak in 2/2 positive GISTs (100%). In hemangiosarcomas, positivity was detected only in tumours of splenic origin, with the malignant endothelial cells of both vasoformative and solid areas being extensively positive. Positivity in liposarcoma was revealed only in the dedifferentiated component. Regarding GIST, positive staining was observed only in the hypercellular subtype. Figures showing desmin expression in myxosarcoma, undifferentiated sarcoma, dedifferentiated liposarcoma, and GIST are available in the supplement.

3.4. Analysis of Keratin AE1/AE3 in Malignant Mesenchymal Tumours (Table 3)

Although we found keratin AE1/AE3 in some tumour types analysed, this does not seem to be a diagnostically useful finding. Positivity was detected in 6/62 tumour samples (10%). Keratin AE1/AE3 was found only among soft tissue sarcomas of the skin, subcutaneous tissue and deep soft tissue. Specifically, 2/8 myxosarcomas (25%) (Figure 4E), and 4/8 undifferentiated pleomorphic sarcomas (50%) (Figure 4F) were moderately positive (all remaining types of malignancies analysed were negative). The extent of staining was classified as 2+ in 1/4 positive undifferentiated pleomorphic sarcomas (25%) and as 3+ in 3/4 positive undifferentiated pleomorphic sarcomas (75%) and 2/2 positive myxosarcomas (100%). The intensity of staining in individual cases varied between tumour areas, with a predominance of moderate intensity.

3.5. Analysis of Keratin 8/18 in Malignant Mesenchymal Tumours (Table 3)

Similar to keratin AE1/AE3, we do not consider the presence of keratin 8/18 in some mesenchymal tumours to be diagnostically useful. Positivity was detected in 17/62 tumour samples (27%). Keratin 8/18 positivity was detected in 7/10 fibrosarcomas (70%) (Figure 5A), 3/8 myxosarcomas (38%), 5/8 undifferentiated pleomorphic sarcomas (63%), 0/6 undifferentiated round cell sarcomas (0%), 2/6 dedifferentiated liposarcomas (33%) (Figure 5B), 0/12 leiomyosarcomas (0%), 0/4 hemangiosarcomas (0%), and 0/8 GISTs (0%). The extent/intensity of staining was classified as 1+/moderate in 2/7 positive fibrosarcomas (29%), as 1+/strong in 1/5 positive undifferentiated pleomorphic sarcomas (25%), as 2+/moderate in 2/7 positive fibrosarcomas (29%) and 2/2 positive dedifferentiated liposarcomas (100%), as 2+/strong in 3/7 positive fibrosarcomas (43%), and as 3+/moderate in 3/3 positive myxosarcomas (100%) and 4/5 positive undifferentiated pleomorphic sarcomas (75%). In dedifferentiated liposarcoma, the positivity was detected only in the well‐differentiated component. Figures showing keratin 8/18 expression in myxosarcoma and undifferentiated sarcoma are available in the supplement.

FIGURE 5.

Immunohistochemical expression of keratin 8/18, KIT and SMA in malignant mesenchymal tumours in guinea pigs. All images immunohistochemical staining. (A) Fibrosarcoma, deep soft tissue of the trunk. Focal and moderate immunoreactivity for keratin 8/18 (400× magnification, scale bar 50 µm). (B) Dedifferentiated liposarcoma, subcutis. Focal and moderate immunoreactivity for keratin 8/18 in the component of the well‐differentiated liposarcoma (400× magnification, scale bar 50 µm). (C) Gastrointestinal stromal tumour, caecum. Diffuse and strong immunoreactivity for KIT (100× magnification, scale bar 200 µm). (D) Leiomyosarcoma, uterus. Diffuse and strong immunoreactivity for SMA (400× magnification, scale bar 50 µm). (E) Hemangiosarcoma, skin. Focal and strong immunoreactivity for SMA in endothelial cells (arrows) and discontinuous immunoreactivity in myopericytes. Adjacent small non‐neoplastic vessel with clearly discernible SMA‐negative endothelium is present (asterisk) (400× magnification, scale bar 50 µm). (F) Hemangiosarcoma, spleen. Focal and strong immunoreactivity for SMA in the solid area (400× magnification, scale bar 50 µm). NA of the objective lens: 0.30 for 100× magnification and 0.75 for 400× magnification.

3.6. Analysis of KIT in Malignant Mesenchymal Tumours (Table 4)

TABLE 4.

Summary of immunohistochemical expression of KIT, SMA, S100, and vimentin in malignant mesenchymal tumours analysed.

Tumour type	KIT *	SMA *	S100 *	Vimentin *
Fibrosarcoma	0/10 (0%)	7/10 (70%)	0/10 (0%)	10/10 (100%)
Myxosarcoma	0/8 (0%)	0/8 (0%)	3/8 (38%)	8/8 (100%)
Undifferentiated pleomorphic sarcoma	0/8 (0%)	8/8 (100%)	2/8 (25%)	8/8 (100%)
Undifferentiated round cell sarcoma	0/6 (0%)	6/6 (100%)	0/6 (100%)	6/6 (100%)
Dedifferentiated liposarcoma	0/6 (0%)	3/6 (50%)	3/6 (50%)	6/6 (100%)
Leiomyosarcoma	0/12 (0%)	12/12 (100%)	0/12 (0%)	9/12 (75%)
Hemangiosarcoma	0/4 (0%)	4/4 (100%)	0/4 (0%)	4/4 (100%)
GIST	8/8 (100%)	8/8 (100%)	0/8 (0%)	8/8 (100%)

Abbreviations: GIST, gastrointestinal stromal tumour; SMA, smooth muscle actin.

^*Positive cases.

This study unequivocally demonstrated the utility of KIT in the diagnosis of GISTs. Positivity was detected in 8/62 tumour samples (13%). Of all the different tumour types employed, KIT positivity was detected only in GISTs. All 8/8 cases (100%) were positive in a diffuse and strong staining pattern, regardless of subtype (Figure 5C).

3.7. Analysis of SMA in Malignant Mesenchymal Tumours (Table 4)

Similar to desmin, SMA was present in a number of mesenchymal neoplasms, but only the extensive positivity found in smooth muscle tumours proved to be diagnostically useful. Positivity was detected in 48/62 tumour samples (77%). Among soft tissue sarcomas of the skin, subcutaneous tissue and deep soft tissue, 7/10 fibrosarcomas (70%), 0/8 myxosarcomas, 8/8 undifferentiated pleomorphic sarcomas (100%), 6/6 undifferentiated round cell sarcomas (100%), and 3/6 dedifferentiated liposarcomas (50%) were SMA positive. The extent of staining was classified 1+ in 3/6 positive undifferentiated round cell sarcomas (50%), 3/3 positive dedifferentiated liposarcomas (100%), as 3+ in 7/7 positive fibrosarcomas (100%), 6/8 positive undifferentiated pleomorphic sarcomas (75%) and 2/6 positive undifferentiated round cell sarcomas (33%), and as 4+ in 2/8 positive undifferentiated pleomorphic sarcomas (25%) and 1/6 positive undifferentiated round cell sarcomas (17%). The intensity of staining was strong in 11/14 undifferentiated sarcomas (79%) and moderate in 3/14 undifferentiated sarcomas (21%), 3/3 positive dedifferentiated liposarcomas (100%) and 7/7 positive fibrosarcomas (100%). In cases of positive liposarcoma, only solid regions of the dedifferentiated component were stained. Regarding uterine leiomyosarcomas, 12/12 cases (100%) demonstrated strong and diffuse expression throughout the tumour tissue (Figure 5D). Immunohistochemical analysis revealed 4/4 hemangiosarcomas (100%) as SMA positive. Expression was found in both vasoformative and solid areas. The staining intensity was strong and the extent of staining was classified as 2+ in vasoformative areas (Figure 5E) and as 1+ in solid areas (Figure 5F). In vasoformative areas, it was difficult to distinguish SMA‐positive endothelium from the outer pericyte layer, but meticulous analysis revealed unequivocal positivity in approximately 20% of endothelial lining cells, while the pericyte layer was focally disrupted (discontinuous).

Staining with an anti‐SMA antibody revealed considerable positivity in 8/8 GISTs (100%). The extent of staining was classified as 4+ in the hypercellular subtype and as 3+ in the palisaded‐vacuolated subtype. In the palisaded‐vacuolated subtype, some nodules showed extensive positivity in more than 75% of tumour cells, while others were completely negative or only sporadically positive. The staining intensity was homogeneously strong throughout the hypercellular subtype, whereas alternating zones of moderate and strong intensity were readily detectable in the palisaded‐vacuolated subtype. Importantly, within each category of staining intensity which was considered positive, that is, moderate or strong, slight cell‐to‐cell heterogeneity was observed. These subtle differences ruled out a false‐positive immunoreaction. A figure showing SMA expression in fibrosarcoma, myxosarcoma, undifferentiated sarcoma, dedifferentiated liposarcoma, and gastrointestinal stromal tumour are available in the supplement.

3.8. Analysis of S100 in Malignant Mesenchymal Tumours (Table 4)

The results of our study did not indicate a substantial role for S100 in the diagnosis of the analysed malignancies. Positivity was detected in 8/62 tumour samples (13%). Among soft tissue sarcomas of the skin, subcutaneous tissue and deep soft tissue, both cytoplasmic and nuclear S100 positivity was found only in 3/8 myxosarcomas (38%) (Figure 6A), 2/8 undifferentiated pleomorphic sarcomas (25%) (Figure 6B), and 3/6 dedifferentiated liposarcomas (50%). In contrast, 0/12 uterine leiomyosarcomas, 0/4 hemangiosarcomas, and 0/8 GISTs were S100 positive (0%). The extent of staining in all S100 positive myxosarcomas and undifferentiated pleomorphic sarcomas was classified as 3+ and 4+, respectively. Staining intensity was strong in 3/3 myxosarcomas (100%) and moderated in 2/2 positive undifferentiated pleomorphic sarcomas (100%). In dedifferentiated liposarcoma, the positivity was detected only in the well‐differentiated component, with the extent/intensity of staining being 4+/strong. Figure showing S100 expression in dedifferentiated liposarcoma is available in the supplement.

FIGURE 6.

Immunohistochemical expression of S100 and vimentin in malignant mesenchymal tumours in guinea pigs. All images immunohistochemical staining. (A) Myxosarcoma, skin. Multifocal and strong immunoreactivity for S100 (200× magnification, scale bar 100 µm). (B) Undifferentiated pleomorphic sarcoma, deep soft tissue of the extremity. Multifocal and moderate immunoreactivity for S100 in neoplastic cells and strong positivity in adipocytes (arrow) serving as a positive internal control (400× magnification, scale bar 50 µm). (C) Leiomyosarcoma, uterus. Multifocal and weak immunoreactivity for vimentin in neoplastic cells (asterisk) and strong positivity in lesional vessels (arrow) serving as a positive internal control (400× magnification, scale bar 50 µm). (D) Leiomyosarcoma, uterus. Diffuse and strong immunoreactivity for vimentin (400× magnification, scale bar 50 µm). NA of the objective lens: 0.50 for 200× magnification and 0.75 for 400× magnification.

3.9. Analysis of Vimentin in Malignant Mesenchymal Tumours (Table 4)

The considerable amount of vimentin found in all types of malignancies analysed does not indicate its diagnostic utility in soft tissue pathology in guinea pigs. Positivity was detected in 59/62 tumour samples (95%). Except for leiomyosarcomas, all 50/50 remaining mesenchymal tumours analysed were vimentin positive in a diffuse and strong staining pattern (100%). Of the 12 leiomyosarcomas analysed, 3/12 (25%) were negative and 9/12 (75%) were diffusely positive (score 4+). In vimentin negative cases, staining was focal/multifocal with only weak intensity (Figure 6C). In 2/9 positive leiomyosarcomas (22%), the staining intensity was strong (Figure 6D). The remaining 7/9 positive cases (78%) demonstrated heterogeneous staining intensity with a predominance of moderate intensity over weak intensity. Figures showing vimentin expression in fibrosarcoma, myxosarcoma, and undifferentiated sarcoma are available in the supplement.

3.10. Summary of the Diagnostic Utility of Individual Markers

CD34 is useful in the diagnosis of hemangiosarcomas. KIT and CD34 are useful in the diagnosis of GISTs. Strong and diffuse expression of desmin and SMA is useful in the diagnosis of leiomyosarcomas. Positivity of various markers in fibrosarcoma, myxosarcoma, and undifferentiated sarcoma is of no diagnostic utility.

4. Discussion

There is a lack of research in veterinary medicine dealing with diagnostic immunohistochemistry in rodents. In our study, vimentin was diffusely and strongly expressed in all analysed tumours except some leiomyosarcomas. Thus, the specificity for vimentin was very low and did not exceed 20% in any of the malignancies employed. One important finding of the present study is that vimentin does not allow to distinguish between different mesenchymal malignancies in guinea pigs.

Our study provides valuable insights into to immunohistochemistry of undifferentiated sarcomas. Considerable amounts of vimentin and muscle markers and lower levels of CD34, keratins, and S100 were detected in these tumours. Given these results, it is important to warn against possible overestimation of marker positivity in undifferentiated sarcoma. Our study points to the fact that a large proportion of undifferentiated sarcomas in guinea pigs could actually represent poorly differentiated sarcoma derived from muscle tissue. However, undifferentiated sarcoma lacks a specific immunophenotype, as focal expression of various markers can be observed in this tumour type (Thway and Fisher 2021). For this reason, SMA or desmin positivity alone does not provide convincing evidence of muscle differentiation in undifferentiated sarcoma and does not justify reclassifying undifferentiated sarcoma as pleomorphic leiomyosarcoma or rhabdomyosarcoma. Finally, we would like to emphasise that the role of immunohistochemistry in the diagnosis of undifferentiated sarcoma is an ancillary one, primarily serving as a means to exclude other pleomorphic tumours.

Another important finding of the present study is the need for proper interpretation of expression patterns of muscle markers, especially SMA, in the diagnosis of spindle cell mesenchymal tumours. We have shown that SMA can label a substantial proportion of fibrosarcoma and GIST tumour cells, but unlike leiomyosarcoma, it never stains 100% of the neoplastic population. Although our results indicate immunohistochemical overlap of SMA between fibrosarcoma and GISTs, which demonstrated multifocal and moderate to strong staining pattern, and leiomyosarcoma, which showed diffuse and strong staining pattern, proper evaluation of the SMA staining pattern may serve as a helpful diagnostic tool in differentiating these morphologically similar tumours. This is an important finding because extensive SMA staining in spindle cell tumour with fascicular growth does not equate to a diagnosis of leiomyosarcoma. Thus, the sensitivity and specificity of the diffuse and strong staining pattern for desmin and SMA in differentiating leiomyosarcoma from its spindle cell mimics in our study reached 100%.

A novel finding of our study is the demonstration of positivity of muscle markers in hemangiosarcomas. In cutaneous tumours, both endothelial cells and pericytes were labelled for SMA. Under normal circumstances, myopericytes are the only cell type in the wall of small non‐neoplastic vessels stained with SMA (Kutcher and Herman 2009). For this reason, SMA positivity in neoplastic endothelial lining is surprising. Furthermore, the detection of SMA‐positive myopericytes in malignant vascular channels may be confusing because angiosarcomas are thought to lack pericytes around neoplastic vessels. Although we are not aware of any specific study where this has been investigated, our findings and those of other authors have demonstrated a nearly continuous layer of SMA‐positive myopericytes around neoplastic vessels (Agrawal et al. 2023). We would also like to point out the diffuse and strong expression pattern of desmin in splenic hemangiosarcomas. This staining pattern can be very confusing in malignant vascular neoplasm exhibiting both vasoformative and solid growth. In this scenario, the correct diagnosis is achieved by staining with one of the endothelial cell markers, for example, CD31, CD34, ERG or FLI1 (Miettinen et al. 1994; Miettinen et al. 2011), which should be the antibodies of first choice in the immunohistochemical panel when tumours with vascular architecture are addressed.

Another important point of the present study is to highlight the potential pitfalls of immunohistochemistry in the diagnosis of dedifferentiated liposarcoma. Similar to human and canine liposarcomas (Hasegawa et al. 2000; LaDouceur et al. 2017), we found focal desmin and SMA positivity in the dedifferentiated component, which could lead to a misdiagnosis such as leiomyosarcoma. Furthermore, co‐expression of SMA with another muscle marker, such as desmin, which has also been reported in both human and canine liposarcomas (Hasegawa et al. 2000; Machado et al. 2025), could also raise a high suspicion for rhabdomyosarcoma as well. The distinction between liposarcoma and rhabdomyosarcoma is of clinical significance in veterinary medicine, especially in dogs, where rhabdomyosarcomas generally exhibit more aggressive behaviour compared to the former. In diagnostically difficult cases, the presence of a well‐differentiated liposarcomatous component, overexpression of MDM2 protein and amplification of 12q14‐q15 serve as very valuable diagnostic tools in resolving the differential diagnosis of dedifferentiated liposarcoma (Saâda‐Bouzid et al. 2015).

Finally, we briefly summarise the findings from our study that appear to be nonspecific. In addition to the already mentioned SMA and desmin positivity in undifferentiated sarcoma, there is positivity for keratin 8/18 in fibrosarcoma and dedifferentiated liposarcoma, and positivity for desmin, keratins, and S100 in myxosarcoma. The question remains whether the positivity of muscle markers in hemangiosarcomas is truly specific. This means whether the SMA antibody employed reacts with tumour‐specific epitopes or binds nonspecifically to cross‐reacting antigens. For the sake of completeness, we will mention possible causes of nonspecific background staining. These include nonspecific protein binding, incomplete removal of paraffin, poorly fixed or necrotic tissue, thick preparation, inappropriately concentrated antibody, endogenous biotin, incomplete rinsing of slides, and too intense chromogen staining (Donato 1999).

5. Conclusion

Our study confirmed the utility of CD34 in the diagnosis of hemangioarcomas, as well as CD34 and KIT in the diagnosis of GISTs. The strong and diffuse staining pattern of SMA and desmin proved to be a valuable tool to differentiate leiomyosarcoma from its spindle cell mimics. Thus, a pattern‐based approach is of diagnostic value when evaluating desmin and SMA immunohistochemistry in malignant mesenchymal neoplasms in guinea pigs. A novel finding is that SMA expression was demonstrated in malignant endothelium. However, it remains unclear whether this positivity is truly specific or is rather due to nonspecific cross‐reacting antigens. Vimentin has not been shown to be a useful marker for distinguishing mesenchymal malignancies in guinea pigs.

Finally, several findings that appear to be nonspecific, such as keratin positivity in fibrosarcoma and dedifferentiated liposarcoma, desmin, keratins, and S100 positivity in myxosarcoma, and keratin, SMA and desmin positivity in undifferentiated sarcoma, point to potential pitfalls in diagnostic immunohistochemistry and emphasise the need for a panel of immunohistochemical stains in the investigation of guinea pig mesenchymal tumours. Further studies investigating the immunoexpression of various markers in mesenchymal tumour pathology in guinea pigs are needed.

Author Contributions

Jiří Lenz: Conceptualisation (lead); writing – original draft preparation (lead); investigation (lead); resources (supporting); methodology (lead). Míša Škorič: Writing – review & editing (lead); investigation (equal); validation (lead); supervision (lead). František Tichý: Writing – review & editing (supporting); project administration (lead); investigation (supporting); formal analysis (lead); resources (lead). Luděk Fiala: Funding acquisition (lead); project administration (equal); visualisation (lead).

Funding

This work was supported by the project 2024/ITA/15 of the University of Veterinary Sciences Brno.

Ethics Statement

All the experimental procedures including tissue sampling were conducted according to the ARRIVE guidelines, U.K. Animals (Scientific Procedures) Act, 1986 and the associated guidelines, EU Directive 2010/63/EU for animal experiments and Guide for the Care and Use of Laboratory Animals. The study was approved by the Ethics Committee of the Faculty of Veterinary Medicine, University of Veterinary Sciences, Brno.

Conflicts of Interest

The authors state that there are no conflicts of interest regarding the publication of this article.

Supporting information

Supporting Material: vms370801‐sup‐0001‐SuppMat.docx

Lenz, J. , Škorič M., Tichý F., and Fiala L.. 2026. “Immunoreactivity for CD34, Desmin, Keratins, KIT, Alpha‐Smooth Muscle Actin, S100, and Vimentin in Malignant Mesenchymal Neoplasms in Guinea Pigs: A Series of 62 Cases From a Single Institution.” Veterinary Medicine and Science 12, no. 2: e70801. 10.1002/vms3.70801

Data Availability Statement

The data analysed in this study are available from the corresponding author on reasonable request.