**2. Results**

## *2.1. Patients and Demographics*

A total of 47 BPS/IC female patients available in the hospital records and that were refractory to lifestyle changes and oral/intravesical therapies (including non-opioid analgesics and anti-depressant drugs and anti-histaminics) were included. All patients received at least one intratrigonal injection of 100 U of OnaBotA. A total of 193 procedures were performed as depicted in Table 1, varying between 1 (10 patients) and 14 (one patient). The proportion of patients that received four or more treatments was 48% (see Table 1).

The mean age of the patients at the time of the first injection was 50.7 ( ±14.5) years. The mean initial VAS score of the cohort was 5.7 ( ±1.7), and the mean follow-up is 8.8 ( ±4.2) years. The cohort demographics are presented in Table 2.


**Table 1.** Number of patients by the number of treatments.

**Table 2.** Cohort demographics.


#### *2.2. Duration of Effect per Injection*

The median interval between the injection and the patient's request for a new injection is graphically shown in Figure 1, and it was 500.5 days (P25: 350; P75: 581), The duration of the effect of each treatment seems to be relatively stable during the follow-up. The median time between the first injection and the request for the second injection was the shortest with a median duration of 390 days (P25: 287; P75: 590). The median intervals for patients' requests for another treatment increased afterwards, ranging between 414 days and 669 days.

To investigate possible predictors of response, three groups of patients were defined. Group A comprised all patients currently in treatment, which included patients with the disease controlled confirmed at a visit or patients who already asked for a new injection). Group B comprised patients who were non-responders to OnaBotA (they could have shown treatment response at an initial phase of the OnabotA program but meanwhile the treatment lost efficacy). Group C included patients lost to follow-up due to non-therapeutic causes.

When comparing the duration of effect between responders and non-responders, no differences were found, as shown in Table 3. We omitted group C in this comparison given the low number of patients and since the loss to follow-up was not related to the response to the toxin injections.

The 10th treatment onward medians were not represented in Figure 1 and Table 3 since only two patients reached that number of procedures.

**Figure 1.** Median duration of effect between injections (*Y*-axis in days of effect duration counting time between an injection and the patient request for a new treatment).


**Table 3.** Median duration of the effect of each injection per injection, per group.

Interestingly, of the nine patients that had a response duration below the 25th percentile after the first injection, only two abandoned the OnabotA program due to a lack of response. The other seven are still in treatment, having more than five injections. Moreover, these patients had in the following injections a duration of effect within the median time. This suggests that the duration of the effect of the first injection should not be used as a predictor of long-term treatment success or failure.

## *2.3. Global Treatment Maintenance*

Given that we could analyze the total number of patients treated with OnaBotA and identify which of them are still in treatment per each treatment, we were able to organize a Kaplan-Meier treatment maintenance graphic. The results are shown in Figure 2.

Notice that the graphic, which is not time-related, shows that more than half of the patients, 53%, had a favorable treatment response for a high number of treatments.

## *2.4. Possible Predictors*

To evaluate the possible predictors, we compared the characteristics from group A and group B patients. The results are shown in Table 4.

**Figure 2.** Kaplan-Meier curve of the cohort maintenance in OnaBotA treatment. The *X*-axis represents the number of injections, and the *Y*-axis represents the relative number of patients that were treated (100%). Each drop on the graphic line represents patients that lost treatment efficacy and the vertical dash along the graphic represents the moment patients were lost to follow-up. A total of 53% remained on treatment.

**Table 4.** Patients' features as possible predictors, per group.


As shown, 25 patients remain in treatment and 17 were non-responders to OnabotA. Notice that only one patient in group A had just one treatment (the patient has no symptoms after an injection carried out 15 months ago, with occasional flares easily managed with simple conservative measures).

The initial VAS score and the duration of the effect of the first treatment were not statistically different between responders (group A) and non-responders (Group B). The overall time between treatment and the subsequent request for reinjection was numerically inferior in the long-term responders in group A when compared to Group B, although the difference was not statistically significant (*p* = 0.72).

## *2.5. Adverse Events*

In terms of adverse events, and after analyzing all of the procedures, three types of complications were reported: lower urinary tract infection, straining during micturition, and acute urinary retention with a high post-voiding residual with the need to initiate clean intermittent catheterization (CIC). The results are presented in Table 5.

In 71 procedures, there was no specific mention of adverse events, and thus the adverse events were classified as omitted. Of those procedures with information available, 58 procedures had no adverse event, simple urinary tract infection UTI was recorded in 36 cases, and UTI with symptoms of straining was reported in 13 post-op procedures. Straining without UTI occurred after 14 procedures and straining with incomplete voiding, with the need for CIC, occurred in only one procedure (0.1%). No upper urinary tract infection was recorded.

**Table 5.** Total frequency of adverse events.

