**3. Discussion**

This long follow-up pilot study demonstrated that repeated intravesical instillations of Sapylin after BoNT/A injection could remarkably improve lower urinary tract symptoms and increase the bladder voided volume in IC/BPS patients with tolerable safety, for a long-term effective duration, which was apparently better than BoNT/A injection alone.

It is proposed that IC/BPS might be induced by the interaction among nervous, immune and endocrine factors [1]. The glycosaminoglycan layer protects the bladder mucosa as a chemical barrier against urine. When this layer is defective, it cannot protect the bladder mucosa from infiltrated urine that would induce submucosal inflammation, stimulate persistent sensory nerve hyperactivity and upregulate the urothelium permeability, contributing to urinary frequency and pain [3,4,20].

Previous studies demonstrated that BoNT/A was an effective therapy for IC/BPS; it was recommended by the American Urological Association [21] and the East Asian Urological Association [1]. BoNT/A can inhibit the release of neurotransmitters and neuropeptides (nerve growth factor, calcitonin gene-related peptide and substance P et al.) that regulate pain and inflammation from nerve fibers in the bladder wall and the urothelium to reduce neurogenic inflammation, alleviate neural hypersensitivity and inhibit bladder muscle contraction, which finally improves lower urinary tract symptoms [6]. In 2004, Smith et al. [22] studied 13 IC/BPS patients injected with BoNT/A into the trigone and bladder base, resulting in 69% of the patients seeing a significant improvement in pain. Later, Giannantoni et al. submucosally injected 200 U of BoNT/A into the trigone and bladder floor of 14 patients with IC/BPS. Among them, 12 patients (85.7%) reported subjective improvement at the 1- and 3-month follow-ups; however, the duration only lasted 3 months [23]. In the same group of patients, 13 were followed up with repeated BoNT/A injections for 2 years. A mean of 4.8 ± 0.8 injections were administered per patient, and the mean interval between two consecutive injections was 5.25 ± 0.75 months [24]. Another study reported that at the 5-month follow-up, the beneficial effects persisted in 26.6% of cases and at 12 months after treatment, pain recurred in all the patients [25]. These clinical studies show that the therapeutic duration of BoNT/A in IC/BPS patients is around 3–6 months. Similarly, our present study showed that the response rates of those participants who accepted only a BoNT/A submucosal injection at 3, 6, 9 and 12 months post-injection were 67.9%, 28.6%, 7.1% and 0%, respectively. Interestingly, a randomized comparative study enrolled 34 patients with refractory IC/BPS who were injected with 100 U BoNT/A, mainly into the suburothelial layer [26]. The response rate was 73.5% at 1 month, 58.8% at 3 months, 38.2% at 6 months and 20.6% at 12 months. However, in that study, patients who reported "slightly improved," "improved" or "remarkably improved," were considered as the responders to treatment. Further, the treated population might also differ between that study and ours.

Immunotherapy is a reasonable option for IC/BPS [27]. Many researchers found that Sapylin also had immunotherapeutic effects on bladder cancer by initiating marked lymphocytic infiltration around the tumor cells and inhibiting their growth [16,18]. These results suggested that Sapylin might regulate bladder immune and inflammatory responses.

In the present study, our results also demonstrated that the response duration of those who received BoNT/A with Sapylin was 27.5 months on average. This long-lasting effect might be attributable mainly to the repeated intravesical treatment of Sapylin. The mechanisms of Sapylin in the treatment of IC/BPS, however, were not investigated in this study. Interestingly, Kong et al. [28] demonstrated that Sapylin could stimulate the body to secrete a variety of cytokines to accelerate wound healing by promoting endothelial cell proliferation, migration and angiogenesis and increasing fibroblast migration and collagen deposition. These results are similar to an animal study [19]. These promising discoveries have encouraged us to explore whether Sapylin could induce immune responses in the bladder to stimulate the proliferation of bladder epithelial cells and inhibit the expression of those inflammatory factors initiated by infiltrated urine.

Our study has several limitations. This was a retrospective, preliminary and singlecenter pilot study. Although the present study has biases, it was designed as a pilot study to confirm a novel combined therapy for IC/BPS patients. Moreover, we did not directly compare BoNT/A plus Sapylin with other traditional treatments for IC/BPS. Moreover, post-void residual urine and other side effects need to be investigated comprehensively. Finally, we failed to compare the therapeutic effect between BoNT/A plus Sapylin and Sapylin alone. Further research comparing BoNT/A injection plus Sapylin instillation, BoNT/A alone, or Sapylin alone in the treatment of IC/BPS is warranted with a large, multicenter, randomized, placebo-controlled trial.
