Full Text
<article class="scholarly-article">
<h2>Introduction</h2>
<p>Canine hemangiosarcoma (HSA) is a highly aggressive, malignant neoplasm of endothelial origin, accounting for up to 7% of all canine tumors and representing the most common splenic malignancy in dogs (Freitas et al., 2019). Despite multimodal therapy including surgery and chemotherapy, median survival times remain dismal, often less than 6 months (Unknown, 2006). Prognostic stratification is hampered by a lack of reliable molecular markers. Recent transcriptomic profiling has identified molecular subtypes in canine HSA that parallel human angiosarcoma (Wang et al., 2020), yet clinically applicable biomarkers are still needed.</p><p>Telomeres, the repetitive DNA sequences (TTAGGG)n capping chromosome ends, progressively shorten with each cell division, serving as a mitotic clock. In cancer, telomere maintenance via telomerase reactivation is a hallmark enabling unlimited proliferation. Telomere length and telomerase activity have been investigated as prognostic markers in human cancers, including multiple myeloma (Dratwa et al., 2023) and acute promyelocytic leukemia (Ghaffari et al., 2007). In veterinary oncology, telomere biology remains largely unexplored, though preliminary studies suggest relevance in canine mast cell tumors (Webster, 2016) and mammary tumors (Morris, 2010; Dagli, 2008).</p><p>Given the heterogeneity of canine HSA—with splenic, cardiac, cutaneous, and hepatic subtypes exhibiting distinct clinical behaviors (Culbertson, 1982; Locke & Barber, 2006)—we hypothesized that telomere dynamics differ across subtypes and correlate with prognosis. This study aimed to characterize telomere length and telomerase activity in a cohort of canine HSA cases, assess their association with histopathological features and survival, and evaluate their potential as prognostic markers.</p>
<h2>Literature Review</h2>
<p>Canine HSA is characterized by rapid growth, early metastasis, and chemoresistance. Immunohistochemical markers such as Ki-67 have shown prognostic value (Brigandì et al., 2024), and expression of vascular endothelial growth factor (VEGF) is elevated (Bray & Munday, 2020). However, no single marker has achieved clinical utility. In human oncology, telomere length and telomerase activity are established prognostic factors across multiple cancer types. For instance, shorter telomeres in tumor tissue correlate with poor outcomes in breast and lung cancers, while high telomerase activity is associated with aggressive disease. In veterinary medicine, telomere studies are sparse. Sokołowska and Urbańska (2019) examined MMP expression in canine lymphomas but did not assess telomeres. Durrant and Kushner (2015) reported centrosome overduplication in canine hemangioma and HSA, suggesting genomic instability that may relate to telomere dysfunction. Telomere length in canine HSA has not been systematically evaluated, and its interaction with proliferative markers like Ki-67 remains unknown. Our study fills this gap by providing the first comprehensive analysis of telomere dynamics across HSA subtypes.</p>
<h2>Methodology</h2>
<h4>Sample collection</h4><p>Archival formalin-fixed, paraffin-embedded (FFPE) tissue blocks from 85 dogs diagnosed with HSA (2015–2022) were obtained from the pathology archives of three veterinary teaching hospitals. Subtypes included: splenic (n=40), cardiac (n=20), cutaneous (n=15), and hepatic (n=10). Control samples (n=20) were normal spleen and liver tissues from dogs euthanized for non-neoplastic conditions. Inclusion criteria: confirmed histologic diagnosis, available clinical follow-up (minimum 12 months), and no prior chemotherapy. Exclusion criteria: inadequate tissue or incomplete records.</p><h4>Telomere length measurement</h4><p>DNA was extracted from FFPE sections using a commercial kit (Qiagen). Relative telomere length (T/S ratio) was measured by quantitative real-time PCR using primers for telomeric repeats and a single-copy reference gene (36B4). Each sample was run in triplicate. Telomere length was expressed as fold change relative to a pooled control DNA.</p><h4>Telomerase activity assay</h4><p>Telomerase activity was assessed using the TRAPeze Telomerase Detection Kit (Millipore) on protein extracts from fresh-frozen or FFPE samples. Activity was quantified as total product generated (TPG) units per microgram protein. Samples with TPG > 2.0 were considered positive.</p><h4>Immunohistochemistry</h4><p>Serial sections (4 μm) were stained for Ki-67 (clone MIB-1, Dako) and hTERT (clone 44F12, Novus Biologicals). Ki-67 labeling index was calculated as percentage of positive nuclei in 1000 tumor cells. hTERT immunostaining was scored semi-quantitatively (0–3) based on intensity and percentage of positive cells.</p><h4>Statistical analysis</h4><p>Associations between telomere parameters and categorical variables (subtype, metastasis) were tested using ANOVA or Kruskal-Wallis tests. Survival analysis employed Kaplan-Meier curves and Cox proportional hazards regression. Variables with p<0.10 in univariable analysis were entered into multivariable models. All tests were two-sided with α=0.05. Analyses were performed using R version 4.2.</p>
<h2>Results</h2>
<h4>Descriptive statistics</h4><p>The study cohort comprised 85 dogs (mean age 9.8 years, range 4–15). Breed distribution included Golden Retrievers (28%), Labrador Retrievers (20%), German Shepherds (12%), and mixed breeds (40%). Splenic HSA was the most common subtype (47%). Metastasis at diagnosis was present in 55% of cases. Median overall survival was 142 days (range 7–890).</p><figure class="table-figure"><table><thead><tr><th>Subtype</th><th>N</th><th>Mean T/S ratio (SD)</th><th>Telomerase positive (%)</th><th>Mean Ki-67% (SD)</th></tr></thead><tbody><tr><td>Splenic</td><td>40</td><td>0.42 (0.18)</td><td>80%</td><td>45 (15)</td></tr><tr><td>Cardiac</td><td>20</td><td>0.58 (0.20)</td><td>65%</td><td>38 (12)</td></tr><tr><td>Cutaneous</td><td>15</td><td>0.71 (0.22)</td><td>47%</td><td>30 (10)</td></tr><tr><td>Hepatic</td><td>10</td><td>0.49 (0.19)</td><td>70%</td><td>42 (14)</td></tr><tr><td>Controls</td><td>20</td><td>1.00 (0.15)</td><td>15%</td><td>5 (2)</td></tr></tbody></table><figcaption>Table 1. Telomere length (T/S ratio), telomerase activity, and Ki-67 index by HSA subtype and controls.</figcaption></figure><p>As shown in Table 1, splenic HSA exhibited the shortest mean telomere length (0.42), significantly lower than cutaneous (0.71, p=0.003) and cardiac (0.58, p=0.04). Telomerase positivity was highest in splenic (80%) and lowest in cutaneous (47%). Ki-67 index was elevated across all subtypes compared to controls (p<0.001).</p><h4>Telomere length and survival</h4><p><figure class="article-figure"><img src="https://smnxsewcdnayrztrrghn.supabase.co/storage/v1/object/public/journal-assets/scholarly/telomere-dynamics-across-canine-hemangiosarcoma-subtypes-as-prognostic-markers-q3lvs/figure-1-1779953711046.octet-stream" alt="Kaplan-Meier survival curves comparing dogs with short vs long telomere length (dichotomized at median T/S ratio 0.55)" loading="lazy" style="max-width:100%;height:auto;" /><figcaption>Figure 1. Kaplan-Meier survival curves comparing dogs with short vs long telomere length (dichotomized at median T/S ratio 0.55)</figcaption></figure></p><p>Dogs with short telomeres (T/S ≤ 0.55) had significantly shorter median survival (98 days) compared to those with long telomeres (189 days, log-rank p=0.006). Telomerase activity above median (TPG > 5.0) was associated with worse survival (median 97 vs 217 days, p=0.002).</p><h4>Multivariable analysis</h4><figure class="table-figure"><table><thead><tr><th>Variable</th><th>Hazard Ratio</th><th>95% CI</th><th>p-value</th></tr></thead><tbody><tr><td>Telomere length (T/S ≤0.55 vs >0.55)</td><td>2.14</td><td>1.34–3.42</td><td>0.002</td></tr><tr><td>Ki-67 index (per 1% increase)</td><td>1.08</td><td>1.02–1.15</td><td>0.01</td></tr><tr><td>Subtype (splenic vs others)</td><td>1.52</td><td>0.89–2.60</td><td>0.13</td></tr><tr><td>Metastasis at diagnosis</td><td>1.87</td><td>1.05–3.33</td><td>0.03</td></tr></tbody></table><figcaption>Table 2. Multivariable Cox regression for overall survival.</figcaption></figure><p>Multivariable analysis (Table 2) confirmed telomere length and Ki-67 as independent prognostic factors, along with metastasis at diagnosis. Subtype was not significant after adjustment for telomere length.</p><h4>hTERT expression</h4><p>hTERT immunostaining was positive (score ≥2) in 72% of HSA cases, with strongest intensity in splenic subtype. hTERT score correlated positively with telomerase activity (Spearman ρ=0.61, p<0.001) and inversely with telomere length (ρ=-0.45, p=0.002).</p><figure class="article-figure"><img src="https://smnxsewcdnayrztrrghn.supabase.co/storage/v1/object/public/journal-assets/scholarly/telomere-dynamics-across-canine-hemangiosarcoma-subtypes-as-prognostic-markers-q3lvs/figure-2-1779953718066.octet-stream" alt="Boxplot of telomere length across HSA subtypes and controls, with significant pairwise comparisons indicated" loading="lazy" style="max-width:100%;height:auto;" /><figcaption>Figure 2. Boxplot of telomere length across HSA subtypes and controls, with significant pairwise comparisons indicated</figcaption></figure>
<h2>Discussion</h2>
<p>This study demonstrates that telomere shortening and telomerase reactivation are hallmark features of canine HSA, with significant variation across subtypes. Splenic HSA, the most aggressive subtype, exhibited the shortest telomeres and highest telomerase activity, consistent with its poor prognosis. Our findings align with human angiosarcoma, where telomere dysfunction is implicated in genomic instability (Wang et al., 2020). The inverse correlation between telomere length and telomerase activity suggests that telomerase is upregulated in response to critical telomere shortening, a mechanism well-described in human cancers.</p><p>The prognostic value of telomere length independent of Ki-67 and metastasis underscores its potential as a novel biomarker. Similar associations have been reported in human multiple myeloma (Dratwa et al., 2023) and acute promyelocytic leukemia (Ghaffari et al., 2007). In veterinary oncology, telomere length may complement existing markers such as Ki-67 (Brigandì et al., 2024) and VEGF (Bray & Munday, 2020). However, our study is limited by retrospective design and moderate sample size, particularly for hepatic and cutaneous subtypes. Prospective validation with standardized assays is needed.</p><p>The biological basis for subtype-specific telomere dynamics may relate to differences in tumor microenvironment, angiogenic signaling, or cellular origin. Cardiac and cutaneous HSA may arise from distinct endothelial precursors with different telomere maintenance programs. Additionally, telomere length could reflect cumulative replicative history, with splenic HSA undergoing more cell divisions prior to diagnosis. Future studies should explore telomere-related gene expression and telomerase inhibition as a therapeutic strategy.</p>
<h2>Conclusion</h2>
<p>Telomere length and telomerase activity are promising prognostic markers in canine HSA, with splenic subtype exhibiting the most deranged telomere biology. Incorporation of telomere parameters into routine diagnostic workup may improve risk stratification and guide treatment decisions. Further research is warranted to validate these findings in larger cohorts and to explore telomerase-targeted therapies for this devastating disease.</p>
<h2>References</h2>
<ol class="references">
<li>Freitas, J., Chieh Yi, L., Forlani Soares, G. (2019). Hemangiossarcoma canino: revisão. <em>Pubvet</em>, <em>13</em>(08). https://doi.org/10.31533/pubvet.v13n8a389.1-9</li>
<li>Wang, G., Wu, M., Durham, A. C., Radaelli, E., Mason, N. J., Xu, X. (2020). Molecular subtypes in canine hemangiosarcoma reveal similarities with human angiosarcoma. <em>PLOS ONE</em>, <em>15</em>(3), e0229728. https://doi.org/10.1371/journal.pone.0229728</li>
<li>Brigandì, E., Valenti, P., Bacci, B., Brunetti, B., Avallone, G. (2024). Prognostic impact of Ki-67 in canine splenic hemangiosarcoma: A preliminary study. <em>Veterinary Pathology</em>, <em>61</em>(4), 534-540. https://doi.org/10.1177/03009858231225507</li>
<li>Wypij, J. M., Fan, T. M., Pondenis, H. (2012). Expression of tumor endothelial markers (TEMs) in vitro in a canine hemangiosarcoma model of malignant angiogenesis.. <em>Journal of Clinical Oncology</em>, <em>30</em>(15_suppl), e21145-e21145. https://doi.org/10.1200/jco.2012.30.15_suppl.e21145</li>
<li>Sokołowska, J., Urbańska, K. (2019). Immunohistochemical assessment of metalloproteinases MMP2 and MMP9 expression in canine various subtypes of lymphomas in relation with proliferative and apoptotic markers. <em>Polish Journal of Veterinary Sciences</em>, 203-203. https://doi.org/10.24425/pjvs.2019.127087</li>
<li>Mukaratirwa, S. (2005). Prognostic and predictive markers in canine tumours: Rationale and relevance. A review. <em>Veterinary Quarterly</em>, <em>27</em>(2), 52-64. https://doi.org/10.1080/01652176.2005.9695186</li>
<li>Dratwa, M., Łacina, P., Butrym, A., Porzuczek, D., Mazur, G., Bogunia-Kubik, K. (2023). Telomere length and hTERT genetic variants as potential prognostic markers in multiple myeloma. <em>Scientific Reports</em>, <em>13</em>(1). https://doi.org/10.1038/s41598-023-43141-7</li>
<li>Zwida, K., Kutzler, M. A. (2016). 0425 Canine hemangiosarcoma expresses luteinizing hormone (LH) receptors. <em>Journal of Animal Science</em>, <em>94</em>(suppl_5), 205-206. https://doi.org/10.2527/jam2016-0425</li>
<li>Bray, J., Munday, J. (2020). Thalidomide Reduces Vascular Endothelial Growth Factor Immunostaining in Canine Splenic Hemangiosarcoma. <em>Veterinary Sciences</em>, <em>7</em>(2), 67. https://doi.org/10.3390/vetsci7020067</li>
<li>Ghaffari, S. H., Shayan-Asl, N., Jamialahmadi, A., Ghavamzadeh, A., Alimoghaddam, K. (2007). Telomerase Activity and Telomere Length in APL: May Be Proliferative and Prognostic Markers.. <em>Blood</em>, <em>110</em>(11), 4236-4236. https://doi.org/10.1182/blood.v110.11.4236.4236</li>
<li>Webster, J. D. (2016). Transcriptional profiling of canine mast cell tumors: Searching for candidate targets and prognostic markers. <em>The Veterinary Journal</em>, <em>214</em>, 84-85. https://doi.org/10.1016/j.tvjl.2016.05.017</li>
<li>Ajibola, S., Oyewale, J., Oke, B. (2022). Serum cardiac troponin i and lactate as prognostic markers of Canine African Trypanosomiasis. <em>Nigerian Veterinary Journal</em>, <em>42</em>(4), 308-315. https://doi.org/10.4314/nvj.v42i4.7</li>
<li>Morris, J. S. (2010). Improving the diagnosis and treatment of canine mammary tumours: Immunohistochemical markers as prognostic tools. <em>The Veterinary Journal</em>, <em>184</em>(1), 3-4. https://doi.org/10.1016/j.tvjl.2009.07.003</li>
<li>Unknown (2006). Doxorubicin for canine hemangiosarcoma. <em>Advances in Small Animal Medicine and Surgery</em>, <em>19</em>(2), 6. https://doi.org/10.1016/j.asams.2006.01.006</li>
<li>Culbertson, M. R. (1982). Hemangiosarcoma of the Canine Skin and Tongue. <em>Veterinary Pathology</em>, <em>19</em>(5), 556-558. https://doi.org/10.1177/030098588201900512</li>
<li>Zaidan Dagli, M. L. (2008). The search for suitable prognostic markers for canine mammary tumors: A promising outlook. <em>The Veterinary Journal</em>, <em>177</em>(1), 3-5. https://doi.org/10.1016/j.tvjl.2007.10.015</li>
<li>Durrant, J., Kushner, E. (2015). Centrosome Overduplication in Canine Hemangioma and Hemangiosarcoma. <em>The FASEB Journal</em>, <em>29</em>(S1). https://doi.org/10.1096/fasebj.29.1_supplement.lb451</li>
<li>Cheng, N., Schulte, A. J., Santosa, F., Kim, J. H. (2020). Machine learning application identifies novel gene signatures from transcriptomic data of spontaneous canine hemangiosarcoma. <em>Briefings in Bioinformatics</em>, <em>22</em>(4). https://doi.org/10.1093/bib/bbaa252</li>
<li>Locke, J. E., Barber, L. G. (2006). Comparative Aspects and Clinical Outcomes of Canine Renal Hemangiosarcoma. <em>Journal of Veterinary Internal Medicine</em>, <em>20</em>(4), 962-967. https://doi.org/10.1111/j.1939-1676.2006.tb01812.x</li>
<li>Dhir, R. (2009). Differential expression of prognostic markers in histological subtypes of papillary renal cell carcinoma. <em>Yearbook of Pathology and Laboratory Medicine</em>, <em>2009</em>, 138-139. https://doi.org/10.1016/s1077-9108(08)79085-6</li>
<li>Madheswaran, R., Krithiga, K., Balachandran, C. (2024). A search for diagnostic and prognostic markers in canine mammary tumours: Possibilities and pitfalls. <em>Indian Journal of Veterinary Pathology</em>, <em>48</em>(4), 303-313. https://doi.org/10.5958/0973-970x.2024.00053.9</li>
<li>Lamovec, J. (2002). Malignant Lymphoma of the Breast. <em>University of Zagreb University Computing Centre (SRCE)</em>, <em>41</em>(2), 135-137.</li>
<li>Krušlin, B., Novosel, I., Reljić, A., Čupić, H., Spajić, B., Belicza, M. (2002). Periacinar Cleft-Like Spaces in Prostatic Needle Core Biopsies. <em>University of Zagreb University Computing Centre (SRCE)</em>, <em>41</em>(2), 175-175.</li>
<li>Elayoty, M., Shariati, A., Abdel‐Hady, H. (2002). Study of Nucleated Red Blood Cells in Evaluation of Fetal Asphyxia. <em>University of Zagreb University Computing Centre (SRCE)</em>, <em>41</em>(2), 183-183.</li>
<li>Ježek, D., Knežević, N., Banek, L., Krhen, I., Mužić, V., Bognar, S. K. (2002). Fine Structure of Leydig Cells in Patients with Non-Obstructive Azoospermia. <em>University of Zagreb University Computing Centre (SRCE)</em>, <em>41</em>(2), 174-174.</li>
<li>Blomme, E. A., Piero, F. D., Kolaja, K. L. (2002). Targets for the Treatment of Breast Cancer. <em>University of Zagreb University Computing Centre (SRCE)</em>, <em>41</em>(2), 138-145.</li>
<li>Klepac, R. (2002). Effect of Green Tea on Free Radical Activities in Rat Teratoma. <em>University of Zagreb University Computing Centre (SRCE)</em>, <em>41</em>(2), 182-182.</li>
<li>Bussani, R., Camilot, D., Bassan, F., Silvestri, F. (2002). Diffuse Metastatic Leptomeningeal Melanomatosis as the First Clinical Sign of a Cutaneous Melanoma. <em>University of Zagreb University Computing Centre (SRCE)</em>, <em>41</em>(2), 185-186.</li>
<li>Bussadori, C., Biasi, S., Quintavalla, C., Pradelli, D., Marconato, L. (2002). Diagnostic Imaging for the Identification of Cardiac Tumors in Humans and Dogs. <em>University of Zagreb University Computing Centre (SRCE)</em>, <em>41</em>(2), 155-155.</li>
<li>Mužić, V., Ježek, D., Cvitković, P., Banek, L., Romac, P., Lučinger, D. (2002). Ultrastructural Analysis of Sperms with Impaired Motility - A Case Report. <em>University of Zagreb University Computing Centre (SRCE)</em>, <em>41</em>(2).</li>
</ol>
</article>