Therapeutic effects of axitinib, an anti-angiogenic tyrosine kinase inhibitor, on interstitial cystitis – Scientific Reports

In present study, we reviewed the expression of VEGF, PDGF, and their receptors in the bladder tissue of IC/BPS patients and compared them with those of healthy controls. IC patients had higher protein expression of VEGFR-2, PDGF-B, and PDGFR-B than controls. Based on these clinical findings, we investigated the therapeutic effects of axitinib, a tyrosine kinase inhibitor, in a HCl-induced cystitis rat model, which exhibits similar abnormalities in bladder function and histology to those of IC patients. In the HCl rat models, axitinib alleviated pain, abnormalities in voiding, and pathological changes through the downregulation of the mRNA expression of VEGF, PDGF, and their receptors, demonstrating its therapeutic potential.

Treatment strategies for IC/BPS have evolved continuously; however, pain management is the mainstay of treatment throughout therapy. In addition, it is recommended to offer a choice of therapeutic options ranging from the least to the most invasive. These stepwise approaches for IC/BPS include behavioral/non-pharmacological therapy, oral medication, instillation, and procedures, including fulguration of Hunner lesion, hydrodistension, intradetrusor botulinum toxin injection, neuromodulation, and major surgery⁵. However, the level of evidence and recommended strengths are mostly limited, except for those of behavioral/non-pharmacological therapy. Thus, there is still no definite treatment for IC/BPS.

The intractable feature of IC/BPS has stimulated robust attempts to investigate alternative therapeutic strategies including intravesical instillation of platelet-rich plasma^13,14 and injection of stem cells into the bladder¹⁵. Stem cells facilitate the regeneration of damaged tissue by direct differentiation, self-renewal, and paracrine activities, which recruit various cytokines. However, there are some major issues with stem cell therapy, including determining the type, dosage, and route of administration for maximal therapeutic efficacy and safety in humans, which could require numerous trial and error studies. Testing a pre-existing medication that targets the pathogenesis of IC with a known profile might provide a more efficient and interesting option than experimental preclinical studies and clinical trials.

Angiogenesis plays an important role in the pathogenesis of various chronic inflammatory diseases. Hunner’s lesions, which distinguishes IC from BPS, are classically defined as “a circumscript, reddened mucosal area with abnormal vessels radiating towards a central scar with or without coagulum”¹⁶. This characteristic cystoscopic finding suggests that IC may be associated with angiogenesis. This is why we decided to study axitinib, a tyrosine kinase inhibitor with a high affinity for VEGFR 1–3, which are known to regulate angiogenesis. Previous studies investigating the angiogenic components of IC bladders reported increased expression of VEGF⁶. In addition, high expression of CD31 was reported to correlate with the severity of urinary frequency and bladder pain⁷. Another study suggested that blood perfusion is decreased in IC, especially during the filling phase, and hypoxia-inducible factor-1, a transcriptional mediator of VEGF, is also increased in IC bladders¹⁷.

VEGF mainly contributes to angiogenesis and lymphangiogenesis; however, it also plays a key role in bladder inflammation by promoting nerve plasticity. VEGF and its isoforms primarily act through the tyrosine kinase receptors VEGFR-1 and VEGFR-2, with VEGFR-2 being the most characterized VEGFR. The binding of VEGF to VEGFR typically results in angiogenesis (blood vessel formation) and endothelial cell proliferation until the target organ receives enough oxygen. However, in pathological conditions, VEGF secretion is not restricted resulting in the formation of immature blood vessels that are hemorrhagic and leaky⁸. A recent preclinical study reported that blockade of VEGF/VEGFR-2 signaling in rat model of acute and chronic cyclophosphamide (CYP)-induced cystitis resulted in increased bladder capacity and voided volume¹⁸. Similarly, imatinib, a tyrosine kinase inhibitor that inhibits PDGFR-A, -B, and C-kit, alleviated the altered expression of such receptors in rat bladders with CYP-induced cystitis¹⁹.

The main difference between our human IC bladder and previous studies was the expression level of VEGF. No significant difference in VEGF positive staining was observed between IC patients and controls, while the expression of VEGF was more significant in previous reports. We assume the discrepancy between references and our study results from the characteristics of the control patients and differences in disease duration. First, the control patients from the study by Kiuchi et al.⁶, were bladder cancer patients while controls in our study were healthy non-cancerous, and non-IC/BPS patients. It is impossible to perform direct one-to-one comparison between bladder tissue of non-cancerous patients and normal looking lesion of the bladder cancer patients. There is a report on VEGF expression in bladder cancer patients suggested that bladder cancer tissue has significantly higher VEGF mRNA levels than that of adjacent normal mucosa²⁰. However, the VEGF expression level of tissues in bladder cancer patients are diverse; one study reported that negative VEGF expression was observed in 7.1% of bladder cancer tissue and the expression level of VEGF varied²¹. As oncological demographics of the control group is unknown, further studies on VEGF expressions in various lesions of bladder cancer patients and their direct comparison with those of non-cancerous patients are needed. Meanwhile, the control group of study by Furuta et al.⁷, consists of female patients with benign etiology such as urinary incontinence. There is a report that expression of VEGF was significantly higher among males in the age group of 50 years or older²². The male to female ratio of our control group is 2:3. The included male patients could have levelled-up the VEGF expression of controls.

The expression level of VEGF and its isoforms also differed between our human IC bladder and the HCl rat model. We assume this is due to the gap between the bladder injury (symptom onset) and histological evaluation. In animal models, bladder tissue was uniformly evaluated 1 week after HCl instillation, while the median gap between symptom onset and bladder biopsy was 16.3 months in the human IC and BPS groups. Acute chemical-induced bladder injury in the HCl model might have resulted in more extensive expression of VEGF, PDGF-B, and the associated receptors to restore the damaged tissue. Conversely, longer symptom duration (> 6 months) in IC/BPS patients might result in more chronic or stabilized changes in the urinary bladder. The baseline symptomatic demographics and exact diagnostic criteria of included IC patients in above references are unknown. As AUA guideline suggests the symptom duration for more than 6 months, the median gap of 16.3 months between symptom onset and bladder biopsy our study population might reflects more chronic condition.

The expression level of VEGF can be expressed either as intensity or area of positive staining. The problem with using staining intensity is inter-observer differences; thus, results might not be reproducible. In a sub-analysis of human bladder tissue based on four-scale intensities (negative: 0, mild: 1, moderate: 2, strong: 3) with manual reading of slides, there were no differences in the intensity of VEGF, VEGFR-2, PDGF-B, PDGFR-B among groups. However, three patients in the IC group had strong VEGF staining in the plasma cells adjacent to the vessels, which was not observed in the other groups. Interestingly, these patients had symptom (visual analogue scale ≥ 4) or Hunner lesion recurrence at an average of 6.8 months after transurethral resection and coagulation while another two patients without strong VEGF staining of plasma cells had no symptom recurrence at an average follow-up of 11.2 months. This finding supports the previous report that lymphoplasmacytic infiltration is more prominent in IC bladders, and that B-cell abnormalities might be involved in the pathogenesis of IC. In addition, the disease activity of IC and probability of recurrence may be associated with abnormal angiogenesis and inflammatory cell infiltration.

Whether IC and BPS are diseases from different entities remains controversial. Histological evaluation of IC/BPS has revealed that bladder tissues of Hunner-type IC patients present with severe inflammation and urothelial denudation of the entire bladder, whereas non-Hunner-type IC bladder tissues are characterized by fibrosis and increased mast cell infiltration²³. Moreover, immunohistochemical quantification of T-lymphocytes, B-lymphocytes, and plasma cells suggested that IC and BPS are distinct pathological entities⁴. Our finding of differences in the expression of VEGFR-2 and PDGFR-B between the IC and BPS patients also supports the concept that IC and BPS are different diseases with distinct pathophysiologies.

A limitation of this study is that we analyzed bladder tissues from a small number of patients. In addition, our animal model is an acute chemical-induced cystitis rat model, which may not recapitulate the chronic nature of IC/BPS. However, we have demonstrated that axitinib has therapeutic effects in the IC rat model, which could be fundamental for the future application of axitinib in real clinical practice.

In conclusion, oral administration of axitinib improved pain, voiding profiles, and urothelial integrity through the inhibition of angiogenesis in the IC rat model. Axitinib may have potential therapeutic efficacy in IC patients based on the finding that IC bladder tissues also exhibit increased expression of VEGFR-2 and PDGFR-B.