1 Introduction

Interstitial Cystitis/Bladder Pain Syndrome (IC/BPS) and Chronic Prostatitis/Chronic Pelvic Pain Syndrome (CP/CPPS) are debilitating urologic pain disorders affecting millions of Americans [1, 2]. Together, these conditions can be described under the umbrella term, Urologic Chronic Pelvic Pain Syndrome (UCPPS). Few therapeutic advances have been made in the past 25 years and treatment for the wide spectrum of pain and urologic symptoms is often based on trial and error. Past clinical trials for approved pharmacological treatments for IC/BPS in the United States that directly or indirectly target the bladder have shown variable effectiveness. Guidance for both disorders is currently focused on multimodal therapy and self-management [3-5].

The Multidisciplinary Approach to the Study of Chronic Pelvic Pain (MAPP) Research Network was initiated more than a decade ago to advance our understanding of UCPPS and inform the development of new, improved clinical trials [6, 7]. Importantly, Network study results have revealed distinct clinical phenotypes based on the distribution of pain, specifically pelvic-only and widespread pain subgroups. The widespread pain phenotype further showed higher rates of co-morbid Chronic Overlapping Pain Conditions (COPCs) [8]. Neurobiological studies within the MAPP clearly demonstrated altered brain structure and function, enhanced sensitivity to experimental pain, and heightened ex-vivo immunoreactivity in the widespread pain phenotype [9-12]. In addition to this neurobiological evidence, a recent reanalysis of randomized clinical trials for IC/BPS demonstrated that the widespread pain phenotype is associated with treatment efficacy—participants with localized pain showed a better response to pentosan polysulfate and hydroxyzine than participants with widespread pain [13]. Taken together, these findings suggest that widespread pain may be a critical factor to consider when selecting amongst available treatments. To further explore the clinical significance of widespread versus localized pain, the MAPP Network's second phase observational study, the MAPP Symptom Pattern Study (SPS), assessed if the degree of localized versus widespread symptoms in participants correlates with differential response to commonly used therapies.

In the present study we present findings from the SPS-embedded Analysis of Therapies during the Longitudinal Assessment of Symptoms (ATLAS) module [14]. The SPS was a 36-month observational study in which male and female UCPPS participants were assessed through a variety of integrated, urologic and non-urologic measures. The optional ATLAS sub-study provided additional phenotypic characterization for UCPPS participants at the time of initiating one of a group of selected interventions as part of standard care, and again after 12 weeks of treatment. This allowed a direct examination of the response for phenotypic subtypes identified through various pain and urologic symptom measures to different categories of treatments. We hypothesized that participants with widespread pain (and by inference more CNS involvement) would respond preferentially to centrally-directed medication therapies for both pain and urologic outcomes while participants with pain solely localized to the pelvis would respond preferentially to treatments designed to address peripheral pain drivers, such as pelvic floor physical therapy or cystoscopy with hydrodistention.

2 Methods

3 Results

3.2 Widespread Pain Moderates Effects of Peripherally Versus Centrally-Directed Treatment

There was a significant interaction of widespread pain status on outcomes by treatment category for PPS (pain) and BSI (symptom interference) (Table 2). Specifically, the change in PPS from baseline to 12 weeks post-treatment was greater in the high widespread pain group when receiving centrally-directed therapies (−6.21) compared to peripherally-directed therapies (1.25). This equates to a greater than 6-point improvement for the widespread group on the 0–28 PPS scale with centrally-directed therapy (95% CI, −10.39, −2.03), compared to roughly 1 point worsening for this sub-group when receiving peripherally-directed therapies (95% CI, −1.59, 4.09). This average drop is greater than the minimal clinically important difference in PPS (4 points), and represents a more than 40% reduction from pre-treatment levels [19]. The 12 week change in BSI was greater in the low widespread pain group when receiving peripherally-directed therapies (−6.79) compared to centrally-directed therapies (2.01; both p < 0.05). Therefore, the low widespread group receiving peripherally-directed therapies experienced a nearly seven point improvement (i.e., reduction in the BPI score) in bladder interference (95% CI, −12.55, −1.04) on the 0−42 point BPI scale, compared to a roughly 2-point worsening when receiving centrally-directed therapies (95% CI, −4.16, 8.18). This roughly 7-point improvement represents a more than 35% reduction from pre-treatment levels. Figure 1 shows levels of improvement for the BSI by widespread pain status and treatment type. Conversely, there were no significant effects of widespread pain status by treatment category on USS (urologic symptoms) score over the 12 weeks (p > 0.05).

Table 2. Estimated change in outcome measure score from baseline to 12 weeks post-treatment by widespread pain status and treatment category.

	Peripherally-directed, n = 24						Centrally-directed, n = 16
	Low widespread			High widespread			Low widespread			High widespread
	Est.	95% CI		Est.	95% CI		Est.	95% CI		Est.	95% CI		Interaction p value
Δ PPS*	−2.22	−5.28	0.85	1.25	−1.59	4.09	0.34	−2.97	3.66	−6.21	−10.39	−2.03	0.005
Δ USS*	−1.18	−3.67	1.30	−1.18	−3.44	1.08	−0.08	−2.72	2.55	0.86	−2.55	4.28	0.723
Δ BSI†	−6.79	−12.55	−1.04	0.34	−5.06	5.74	2.01	−4.16	8.18	−8.46	−17.21	0.30	0.011

• * Calculated at average age = 40.52, proportion of females = 0.8 and average baseline outcome
• † Calculated at average age = 40.53, proportion of females = 0.816 and average baseline outcome

4 Discussion

The present study provides a proof of concept that the degree of widespread pain in UCPPS (i.e., IC/BPS and CP/CPPS) is associated with differential response to commonly used treatments. Participants with widespread pain showed the most improvement in genitourinary pain when receiving centrally-directed therapies, while participants with localized pain showed the greatest improvement in genitourinary pain and interference symptoms when receiving peripherally-directed therapies. The MAPP Network has demonstrated that widespread pain in UCPPS is strongly associated with the concept of central sensitization/nociplastic pain and these participants are more likely to have COPCs and more severe painful and non-painful symptoms [20], as well as neurobiological changes that suggest entrenched CNS mechanisms promoting pain (discussed further below). These neurobiological differences are likely at the root of why centrally-acting therapies appear to be more beneficial in participants with significant pain outside the pelvic region. These findings support further characterization of widespread/comorbid pain in future clinical trials and research studies to determine if treatment response in UCPPS can be improved by “matching” patient phenotype to appropriate therapies. Indeed, the distinction between localized versus widespread pain may be even more clinically relevant than previously considered, as roughly three quarters of MAPP Network study participants showed substantial degrees of pain outside the pelvic area [20].

The current findings are supported by a recent reanalysis of randomized controlled trials for IC/BPS that considered the potential impact of widespread pain on oral pentosan polysulfate + hydroxyzine, intravesical Bacillus Calmette-Guerin, and oral amitriptyline treatments. This study showed that localized pain responded preferentially to oral PPS + hydroxyzine compared to placebo, while widespread pain did not; intravesical BCG responses were equivalent, and widespread pain responded preferentially to oral amitriptyline versus placebo, while localized pain did not (the low widespread pain group showed a pronounced response to placebo so this finding should be interpreted cautiously) [13]. While this was not a prospective study, the findings support that incorporating widespread pain into clinical decision-making could increase clinical response rates to available treatments.

MAPP Network studies have also revealed neurobiological differences that may underly observed treatment effects. In a MAPP Network brain neuroimaging study participants demonstrated increases in gray matter volume in the supplementary motor area and primary somatosensory (S1) and motor (M1) cortices as a function of the degree of widespread pain [11]. These areas encompass the hip/trunk/pelvic representation in S1/M1 suggesting structural reorganization with potential pathologic consequences. The left S1/M1 area was shown to exhibit higher functional connectivity with the brain's salience network, a constellation of regions involved in determining the importance of stimuli, including those arising within the body. Importantly, this finding was also shown in a cohort of fibromyalgia patients, a condition characterized by widespread pain that has been shown to respond to tricyclic antidepressants and gabapentinoids [11]. As an example, pregabalin has been shown to reduce concentrations of excitatory neurotransmitter in the posterior insula and patterns of functional connectivity associated with clinical pain in fibromyalgia patients [21].

5 Conclusions

The current study, while exploratory, provides critical new support for the importance of targeting therapy to specific symptom profiles for the improvement of treatment effects (i.e., personalized medicine). Here we assessed interventions with commonly accepted localized versus centralized sites of action to test the hypothesis that such therapies would show different efficacy for urologic pelvic pain phenotypes. These findings are consistent with the hypothesis that urologic pelvic pain involves different mechanistic subtypes that can be exploited to provide desperately needed improvement in the efficacy of future trials and ultimately clinical management. It is important to note that urologic symptom outcomes did not show this interaction effect—further research is needed to determine if the apparent differential effects of centrally-acting therapies are confined to pain outcomes.

Limitations and Future Directions: the current study has a small sample size and the participants were all self-selected into treatment categories. Future clinical trials that stratify participants by the presence or degree of widespread pain for treatment selection are needed to further test the significance of these results.

Author Contributions

John Farrar, J. Richard Landis, and Daniel J. Clauw were primarily responsible for the study design. All authors contributed to the drafting and critical revision of the work. Primary analysis was performed by Andrew Schrepf, Kenneth Locke, and J. Richard Landis.

Conflicts of Interest

The authors declare no conflicts of interest.