Background
Low-dose rate brachytherapy using iodine-125 (
125I) seed implantation with or without supplemental external beam radiotherapy (EBRT) is currently a standard treatment for localized prostate cancer (PCa). According to a nationwide prospective cohort study, the Japanese Prostate Cancer Outcome Study of Permanent Iodine-125 Seed Implant (J-POPS), over 25,000 patients with PCa underwent brachytherapy in Japan as of 2014 [
1,
2]. A previous survey investigating the treatment distribution of primary therapy for cT1–2N0M0 PCa at our institute showed a brachytherapy rate of 38% [
3]. While brachytherapy has positive long-term oncological outcomes, there are clinical concerns. These include the incidence of post-treatment adverse effects such as lower urinary tract symptoms (LUTS) and quality of life (QOL) deterioration [
2,
4‐
11].
Worsening LUTSs are one of the most bothersome consequences of brachytherapy; however, acute symptoms tend to mitigate as the baseline is gradually restored within 1 year [
4,
5]. During further follow-up, transient relapse of LUTS is observed in some patients, which Cesaretti et al. termed a “urinary symptom flare” [
12]. Since then, predictive factors of urinary symptom flares have been explored in large cohorts of patients undergoing brachytherapy [
4,
8‐
11]. In previous reports, chronological changes on International Prostate Symptom Score (IPSS) questionnaires were used to evaluate the urinary symptom flare. The results showed that urinary symptom flare incidence was associated with erectile dysfunction, higher baseline IPSS, maximal post-implant IPSS, age, the biologically effective dose (BED), and implementation of supplementary EBRT [
4,
9,
11].
The IPSS has been predominantly utilized as a tool for defining urinary symptom flares, which are termed IPSS flares [
4,
8‐
12]. Bothersome symptoms after radiotherapy seem to mainly be characterized by storage urinary characteristics including frequency, urgency, and nocturia [
13]. The overactive bladder symptom score (OABSS) was developed in 2006 as an evaluation tool for patients with overactive bladder syndrome (OAB) and has been validated in Japanese as well as in other populations [
14‐
16]. The OABSS evaluates relevant symptoms with only 4 questions that cover daytime frequency, nocturia, urgency, and urge incontinence. This questionnaire is simple and quick, and it agrees with corresponding diary variables [
17] and treatment-related improvement with anticholinergic use [
18]. Therefore, the OABSS may be beneficial for evaluating urinary symptom flares after brachytherapy.
To date, no reported studies have investigated urinary symptom flare after 125I brachytherapy in a single cohort by using both the IPSS and the OABSS. In this prospective study, we focused on the predictive factors for urinary symptom flare based on the IPSS and OABSS questionnaires. In addition, we investigated the association between urinary symptom flare and the incidence of prostate-specific antigen (PSA) bounce because research on this issue is limited.
Discussion
The initial worsening of LUTS after implantation has been thoroughly studied by several groups [
2,
5,
8,
10,
11]. We previously reported that the total IPSS and QOL index peak approximately 3 months after implantation and gradually return to the BL scores after 12 months [
5]. The etiology of worsening LUTS revealed that the peak IPSS after seed implantation was associated with the total amount implanted and the dose delivered to the prostatic gland, bladder, and urethra [
24]. The recurrent LUTS exacerbation after a variable asymptomatic period occurred between 6 and 60 months after seed implantation [
12], which is called “urinary symptom flare.” The IPSS flare has been utilized predominantly as a tool for determining urinary symptom flare [
4,
8‐
12]. A previous study evaluated the spectrum of pathophysiology underlying persistent LUTS after seed implantation [
25]. In their cohort, 79% of the patients had overactive symptoms, and 71% had urinary incontinence, while only 44% had obstructive symptoms. A urodynamic study revealed that men undergoing seed implantation had a much higher incidence of detrusor overactivity. Many urologists and general practitioners currently use the IPSS to evaluate the severity of LUTS, decide upon an intervention, and assess the treatment outcome. Total IPSS was reported as unreliable for correctly diagnosing bladder outlet obstruction and OAB [
26]. Additionally, voiding symptoms and storage symptoms do not always directly reflect those dysfunctions [
27]. Since persistent LUTS is largely characterized by storage symptoms, it may be reasonable to use the OABSS to evaluate changes in LUTS after implantation. In the present study, the OABSS as well as the IPSS were accurate, useful, and sensitive to changes during a follow-up survey for patients with localized PCa after seed implantation (Fig.
2b).
Although some reports regarding urinary symptoms after seed implantation have been published, an accepted definition of urinary flare does not exist. An increase in urinary symptom score seems to be useful and practical in clinical settings and academic research. In the first detailed evaluation by Cesaretti et al., when an increase in total IPSS of ≥ 5 points was defined as a urinary flare, 36% of patients were determined to experience a flare [
12]. In the report by Keyes et al., two different definitions of flare, an increase in total IPSS of ≥ 5 and ≥ 8 points, were determined, and significant predictive factors for flare were explored separately [
9]. In this study, an increase in total IPSS of ≥ 12 points and of ≥ 6 points in the OABSS were determined to be clinically significant (Fig.
2). The definition used in this study is thought to be reasonable because it qualifies patients who experience long-lasting LUTS and lower QOL (Fig.
2c and d).
With our definition, approximately 25% of patients were determined to experience urinary symptom flare after the nadir. A multivariate analysis of the possible predictors for the incidence of urinary symptom flare revealed that patients treated by higher BED stratified to <200 Gy2 and ≥200 Gy2 and patients without DM had higher risks of urinary flare (Table
2). A recent paper from Japan demonstrated that high BED was associated with the incidence of IPSS flare (defined as an increase in total IPSS of ≥ 5 points) and urinary toxicity of CTCAE grade 2 or higher, but no significant association was found between BED and the first IPSS resolution [
11]. In the present study, the presence of DM was associated with a lower risk of having an IPSS flare, which was a controversial result [
9,
11]. One possible explanation is that the urinary symptom flare is masked by the generally worse baseline urinary function and symptoms in those with DM as compared to those without the condition [
28]. Neuropathy, which is caused by ischemic change, is known to be a DM-related comorbidity. Insensitivity to inflammation and irritability, which is usually induced by radiotherapy, may be another explanation.
With regards to the PSA bounce after radiotherapy, the predictors of bounce, features that distinguish benign bounce and biochemical failure, and the prognostic impact of bounce have been well studied [
29‐
32]. However, the detailed mechanism underlying PSA bounce remains unclear, as does the mechanism behind a urinary symptom flare. We decided to investigate the association between urinary symptom flare and PSA bounce after implantation because the data are extremely limited [
12]. In our cohort, PSA bounce was observed at a similar rate of approximately 20% in the non-flare and flare groups, showing no significant difference (Additional file
2: Table S2). An additional correlation analysis revealed that the two phenomena did not occur with similar timing. Our findings and the data reported by Cesaretti et al. strongly suggested that the etiologies of bounce and urinary flare are distinct.
Limitations of this study include the relatively small sample size, which lowers the ability to obtain significant results and identify other predictors of urinary symptom flare. This was a single-institution nonrandomized study. Moreover, assessment with the IPSS and OABSS questionnaires was only conducted once a year, starting from the second year after seed implantation. More frequent assessments, such as once every 3 to 6 months, may improve the findings.
Conclusions
To our knowledge, the present study is the first to demonstrate the clinical potential of the OABSS as an assessment tool for urinary symptom flare after seed implantation. We also investigated the predictors of IPSS and OABSS flares and the correlation between PSA bounce and urinary symptom flare. Although this single institution study, with only one surgeon, may not be representative of the experiences in the wider community, the data reflect the current status of LUTS management after implantation. Symptom flare is common and occurs in many patients within 5 years. A future prospective multi-center clinical trial will be needed to develop strategies for employing medications such as alpha-1 adrenoceptor antagonist, anti-cholinergic drugs, and phosphodiesterase type 5 inhibitors such as tadarafil (Cialis, Adcirca) to palliate bothersome symptoms. We believe that combined assessment with the OABSS and IPSS is useful when faced with the decision-making process, as it helps with both the timing and selection of treatment intervention, as well as with tracking of the outcome.