**Sex-Sparing Robot-Assisted Radical Cystectomy with Intracorporeal Padua Ileal Neobladder in Female: Surgical Technique, Perioperative, Oncologic and Functional Outcomes**

**Gabriele Tuderti 1,\*, Riccardo Mastroianni 1,2, Simone Flammia 2, Mariaconsiglia Ferriero 1, Costantino Leonardo 2, Umberto Anceschi 1, Aldo Brassetti 1, Salvatore Guaglianone 1, Michele Gallucci <sup>2</sup> and Giuseppe Simone <sup>1</sup>**


Received: 8 January 2020; Accepted: 14 February 2020; Published: 20 February 2020

**Abstract:** Our aim was to illustrate our technique of sex-sparing (SS)-robot-assisted radical cystectomy (RARC) in female patients receiving an intracorporeal neobladder (iN). From January 2013 to June 2018, 11 female patients underwent SS-RARC-iN at a single tertiary referral center. Inclusion criteria were a cT ≤ 2 N0 M0 bladder tumor at baseline imaging (CT or MRI) and an absence of tumors in the bladder neck, trigone and urethra at TURB. Baseline, perioperative, and outcomes at one year were reported. The median operative time was 255 min and the median hospital stay was seven days. Low-grade Clavien complications occurred in four patients (36.3%), while high-grade complications were not observed in any. Seven patients (63.7%) had an organ-confined disease at the pathologic specimen; nodal involvement and positive surgical margins were not detected in any of the cases. At a median follow-up of 28 months (IQR 14–51), no patients developed new onset of chronic kidney disease stage 3b. After one year, daytime and nighttime continence rates were 90.9% and 86.4% respectively. Quality of life as well as physical and emotional functioning improved significantly over time (all *p* ≤ 0.04), while urinary symptoms and sexual function worsened at three months with a significant recovery taking place at one year (all *p* ≤ 0.04). Overall, 8 out of 11 patients (72.7%) were sexually active at the 12-month evaluation. In select female patients, SS-RARC-iN is an oncologically sound procedure associated with favorable perioperative and functional outcomes.

**Keywords:** bladder cancer; female; intracorporeal neobladder; outcomes; radical cystectomy; robotic; sex-sparing

#### **1. Introduction**

Radical cystectomy (RC) with urinary diversion is the standard treatment for patients with muscle-invasive (MI) and high-risk non-muscle-invasive (NMI) urothelial carcinoma of the bladder and can offer an orthotopic neobladder (ON) diversion if technically and oncologically feasible [1].

Although bladder cancer (BCa) is more frequent among men, it remains the 17th most common cancer in women worldwide, with approximately 74,000 new diagnosed cases each year [2]. Moreover, women present an advanced stage at diagnosis more often, increasing the requirement of RC [3].

*J. Clin. Med.* **2020**, *9*, 577

In female patients, the standard surgical procedure is represented by anterior pelvic exenteration including the removal of the bladder, ovaries, uterus, and anterior vaginal wall [1].

In this setting, when an ON is performed, the procedure can be associated with a considerable rate of voiding symptoms [4,5]. In addition, sexual dysfunction derived from such a highly demolitive surgical procedure is a key concern, especially in younger patients due to a significant impact on health-related quality of life (HRQoL) [6].

The improvement of imaging modalities, an increased knowledge of pelvic structure anatomy and function, and an advancement of surgical techniques have enabled less-destructive methods for treating high-risk BCa.

In this scenario, various types of pelvic-organ-preserving techniques, usually named "sex-sparing", have been proposed [5], aiming at the preservation of neurovascular bundles, vagina, and uterus, combining these techniques in order to optimize sexual and functional results without compromising oncological outcomes.

Functional outcomes of sex-sparing (sex) RC are essentially based on surgical dissection planes, with the sex-sparing approach being associated with the preservation of utero-vaginal hypogastric plexus, while during standard RC only rectal hypogastric plexus is preserved [7].

The Bern group were the first to describe the feasibility of nerve-sparing RC and ON replacement in female patients, highlighting the potential advantages which derive from preserving pelvic reproductive organs and their nervous structures both in terms of continence and urinary retention [7].

However, a recent systematic review aiming at the evaluation of the advantages and disadvantages of sexual-function-preserving RC and ON in female patients underlined the need for further and more robust comparisons between sex and standard RC as existing data are still immature [5].

Notwithstanding, for well-selected patients, sparing female reproductive organs during RC can be an oncologically safe procedure and can provide improved functional outcomes.

Accordingly, despite the widespread use of robot-assisted radical cystectomy (RARC), there is a paucity of data concerning outcomes of sex-RARC with intracorporeal ON (iON) performed in female patients.

In this paper we describe surgical steps of sex-RARC in female patients, highlighting differences with the standard technique and anatomical details of preservation of the inferior hypogastric plexus (IHP) and we report perioperative, pathologic, and functional outcomes.

#### **2. Experimental Section**

#### *2.1. Patients*

Our single-center Institutional-Review-Board-approved BCa database was queried for "Female", "RARC", "iON", and "Sex-sparing". Overall, 11 patients were treated between January 2013 and June 2018, with a minimum one year of follow-up. Inclusion criteria were a cT ≤ 2 N0 M0 bladder tumor at baseline imaging (CT or MRI) and an absence of tumors in the bladder neck, trigone, and urethra at transurethral resection of the bladder tumor (TURB). Exclusion criteria included any contraindication to ON. All subjects gave their written informed consent for inclusion before they participated in the study.

#### *2.2. Surgical Technique*

#### 2.2.1. Sex-Sparing Robot-Assisted Radical Cystectomy

The patient was placed in a steep Trendelenburg position, and a six trocars access was performed as previously described [8].

Sex-RARC was performed replicating the principles of open technique described by Bhatta Dhar et al. [7]. After an incision of the posterior peritoneum up to the round ligament, the ureters were identified and meticulously isolated with a "no-touch" technique. The umbilical artery, uterine artery, superior and inferior vesical arteries, and vaginal branches were carefully prepared bilaterally.

Because the uterus was going to be spared, the peritoneum was incised at the level of the utero-vesical junction in order to deflect the uterus and develop a vesico-vaginal plane between the bladder and the anterior wall of the uterus. The vaginal wall dissection at the cervical level was performed in the anterior plane of the vagina at the 2 and 10 o'clock position in order to preserve the utero-vaginal and pararectal components of the IHP (highlighted in red and green colors respectively, in the video), while in the standard technique the dissection is usually performed dorsolaterally at the 4 and 8 o'clock position, preserving only the pararectal plexus and removing en bloc with the specimen and the utero-vaginal components of the IHP. The superior and inferior vesical arteries and veins were secured with Hem-o-lok clips (Teleflex, Wayne, PA, USA) and transected with LigaSure at their origin from the internal iliac vessels, while the uterine arteries and the vaginal branches directed to the paravaginal tissue were preserved. Both ureters were divided between Weck clips, and margins were sent for frozen sections. Next, the Retzius space was approached. Endopelvic fascia was incised very close to the bladder neck in order to reduce the risk of an accidental injury of neurovascular paraurethral structures, which is crucial for both sexual and continence functionality. The urethra was prepared and a sample was sent for frozen section. Bladder was secured in an endobag and extracted through a 3-cm prepubic incision.

#### 2.2.2. Pelvic Lymph Node Dissection and Intracorporeal Orthotopic Neobladder

A meticulous separate package extending pelvic lymph node dissection (PLND) was performed, including obturator, internal, external, and common iliac nodes. Considering that superior hypogastric plexus (SHP) is usually located just below the aortic bifurcation, ventrally to the sacral promontory, presacral nodes are not removed. Moreover, lymphatic tissue medial to internal iliac arteries which is in close contact with uterine and vaginal vessels and with uterine and vaginal plexus is usually spared.

After RC and PLND, intracorporeal Padua ileal neobladder was performed as previously described [8].

#### *2.3. Outcomes Evaluated*

Collected demographic parameters were age, body mass index (BMI), gender, and American Society of Anesthesiologists (ASA) score. Clinical variables were preoperative eGFR, preoperative hemoglobin (Hgb), and neoadjuvant chemotherapy rate. Surgical outcomes reported consisted of operative time, Hgb at discharge, hospital stay, and complications according to the Clavien–Dindo system [9]. Pathological findings including pT stage, pN stage, histology, lymph node count, and the positive surgical margin status were analyzed. Functional outcomes assessed were the last eGFR, neobladder stones rate, the uretero-ileal strictures rate, and the need for intermittent self-catheterization. Daytime and nighttime continence recovery probabilities were assessed over time. EORTC QLQ-C30 and EORTC QLQ-BLM30 questionnaires were adopted to assess HRQoL and urinary symptoms respectively. Every item measured ranged in a score from 0 to 100. A Female Sexual Function Index (FSFI) questionnaire was adopted for sexual function assessment [10]. Each of the six sexual domains range in score from 0 to 6, with a maximum global score of 36. Questionnaires were administered at baseline, and at 3 and 12 month follow-up.

As supplementary data, we reported preoperative perioperative, pathologic and functional characteristics comparisons of sex-RARC and standard RARC cohorts.

#### *2.4. Statistical Analysis*

Descriptive analyses were used. Frequencies and proportions were reported for categorical variables. Medians and interquartile ranges (IQRs) were reported for continuously coded variables.

The Kaplan–Meier method was performed to report daytime and nighttime continence recovery probabilities. Continence rates were computed at 3, 6, 12 and 18 months after surgery.

Differences between questionnaires' domains scores evaluated at the baseline 3-month, and 1-year follow-up were assessed with the Friedman test.

In the supplementary outcomes, comparison, continuous, and categorical variables were compared with a Student's *t*-test and a chi-square test respectively. The Kaplan–Meier method was performed to compare daytime continence recovery probabilities between sex-RARC and standard RARC cohorts. Continence rates were computed at 3, 6, 12, and 18 months after surgery and the log-rank test was applied to assess any statistically significant differences between the two groups.

All *p*-values < 0.05 were considered statistically significant. Statistical analysis was performed using SPSS v24 (IBM Corp., Armonk, NY, USA).

#### **3. Results**

Baseline and clinical features were reported in Table 1. Median operative time was 255 min (IQR 250–399). The median hospital stay was 7 days (7–12). Low-grade Clavien complications occurred in four patients (36.3%) while high grade complications were not observed. Seven patients (63.7%) had an organ-confined disease at the pathologic specimen; nodal involvement and positive surgical margins were not detected in any case (Table 2).

All patients had a minimum follow-up period of one year. At a median follow-up of 28 months (IQR 14–51), no patient developed a new onset of chronic kidney disease stage 3b. One patient reported a neobladder stone formation, and one patient developed a ureteroileal anastomotic stricture and required robotic reimplantation 18 months following surgery (Table 2).

One-year daytime and nighttime continence recovery probability were 90.9% and 86.4%, respectively (Figure 1a,b). Three patients performed self-catheterization twice a day (early morning and before night rest).

**Figure 1.** (**a**,**b**) Kaplan–Meier analysis reporting daytime and nighttime continence recovery probabilities.


**Table 1.** Baseline and clinical characteristics.


**Table 2.** Perioperative, pathologic, oncologic and functional outcomes.

Concerning the EORTC-QLQ-C30 questionnaire, global health status/quality of life, physical, and emotional functioning items improved significantly over time (all *p* ≤ 0.04), while no differences were observed in any other items evaluated (all *p* ≥ 0.10) (Supplementary Table S1, Figure 2).

According to the EORTC-QLQ-BLM30 questionnaire, specific for BCa, urinary symptoms worsened at three months with a significant recovery at one year (*p* = 0.02). Accordingly, when matching the baseline with 1-year scores, the values were comparable (*p* = 0.08) (Supplementary Table S2, Figure 2).

Finally, the FSFI global score and FSFI domains such as arousal, lubrication, orgasm, satisfaction, and pain worsened over the first three months with a subsequent improvement at one year (all *p* ≤ 0.04). Moreover, comparing baseline vs. 1-year scores, arousal and orgasm domains experienced a complete recovery (*p* = 0.10 and *p* = 0.10, respectively), while lubrication, satisfaction, and pain domains, as well as FSFI global scores, experienced a satisfying improvement but were statistically significantly lower than baseline (all *p* ≤ 0.025) (Supplementary Table S3, Figure 3). Overall, 8 out of 11 patients (72.7%) were sexually active at the 12-month evaluation.

As supplementary analysis, 36 standard RARC patients were compared with the sex-RARC cohort. The two cohorts were homogeneous for all baseline, clinical, and pathological features (all *p* ≥ 0.14) except for age, with sex-sparing patients being significantly younger (47.1 vs. 61.7 years, *p* < 0.001) (Supplementary Tables S4 and S5).

Perioperative complications and hospital stay were comparable between groups (*p* = 0.25 and *p* = 0.67 respectively) (Supplementary Table S5).

**Figure 2.** Graphs showing EORTC-QLQ-C30 and EORTC-QLQ-BLM30 questionnaire items displaying statistical significance according to the Friedman test. (**A**) Quality of Life/Global health status; (**B**) Emotional functioning; (**C**) Physical functioning; (**D**) Urinary symptoms and problems

**Figure 3.** Graphs showing global Female Sexual Function Index (FSFI) and FSFI single domains questionnaire displaying statistical significance according to the Friedman test. (**A**) FSFI; (**B**) Arousal; (**C**) Lubrication; (**D**) Orgasm; (**E**) Satisfaction; (**F**) Pain.

With regard to functional outcomes, no significant differences were observed for the last estimated glomerular filtration rate (*p* = 0.43), neobladder stone formation rate (*p* = 0.93), and 1-year incidence of ureteroileal strictures (*p* = 0.67) (Supplementary Table S5). Daytime continence recovery probability was significantly higher in the sex-sparing cohort (1-year rate 90.9% vs. 74%, log-rank *p* = 0.02) (Supplementary Figure S1).

#### **4. Discussion**

Functional outcomes among women undergoing RC have been poorly addressed in the literature [11]. Urinary function is the most studied issue, although daytime and nighttime continence rates range significantly across studies due to a heterogeneity of definitions for continence, different inclusion criteria, and a lack of questionnaire adoption, as these are omitted in most studies [11]. In addition, Zahran et al. conducted a systematic review aiming to evaluate female sexual dysfunction post RC and urinary diversion, considering it an important predictor of HRQoL post RC. According to the 11 studies included, the most frequently detected sexual disorders were loss of sexual desire and orgasm disorders (49% and 39%, respectively) [12]. Notwithstanding, the authors called for the use of standardized tools in order to properly assess the outcomes of this technique from the patients' perspective and reported poor evidence from the available literature. Moreover, no data were available about RARC in females.

The concept of sex-RC in female patients was first introduced by the Bern team in 2007, when, in select female patients with an absence of invasive cancer at the level of the trigone or dorsolateral side walls of the bladder, they emphasized the functional advantages deriving from the preservation of the utero-vaginal hypogastric plexus, which is usually sacrificed in the standard procedure [7].

These results were corroborated by meticulous cadaveric studies elucidating topographic anatomic details of the nervous autonomic system in women, with their clinical nuances [13,14]. The SHP was identified as a single anatomical complex located below the aortic bifurcation, ventral to the sacral promontory. After the promontory, the SHP divides into right and left hypogastric nerves that more caudally plunge into the inferior pelvic IHP, composed by utero-vaginal, vesical, and rectal plexus.

As expected, preservation of these neural structures has an impact on recovery of urinary continence and on voiding function. Accordingly, data coming from gynecological studies report intrinsic sphincter deficiency resulting from hysterectomy as a consequence of urethral denervation after an extensive pelvic dissection [15,16]. Moreover, a pelvic autonomous nervous system affects all the domains of sexuality, such as sexual desire, arousal, lubrication, orgasm, satisfaction, and post-RC sexual dysfunction, often associated with pain disorders, such as dyspareunia, vulvodynia, and vaginismus, each being a consequence of autonomic and nociceptive nerve injuries, and a shortening or a narrowing of the vagina with an unavoidable negative impact on HRQoL [17].

In the literature, there are few existing series reporting sexual function results after sex-RC, all of them with an open approach and most of them with a small number of patients and without assessment of HRQoL through self-administered standardized questionnaires. Nandipati et al. focused on preservation of the lateral walls of the vagina, in which are embedded nervous fibers directed to the paraurethral tissue, involved in clitoral vascularization. In the small cohort of six women who underwent the sex-sparing approach, 12-month FSFI remained stable, while it declined in the standard RC group [18].

Furthermore, a significant improvement in all domains of the FSFI questionnaire has been reported in 13 sex-sparing RC patients evaluated at Mansoura Urology Department, with daytime and nighttime continence rates of 100% and 92%, respectively [19].

In this context, the EAU MIBCa Guideline Panel recently commissioned a systematic review aiming to assess the effect of sexual-function-preserving surgical techniques on outcomes in women receiving RC and ON substitution for BCa [5]. Sex-sparing approaches were found to be oncologically safe in well-selected patients, with sexual function appearing to be improved among those women undergoing gynecologic organ-preserving and nerve-sparing approaches. Nevertheless, most of the studies analyzed were retrospective and only contained a small number of patients [5]. Hence, according to EAU guidelines, data regarding sex RC in female patients are still considered immature and it is not yet considered a standard treatment, but an option to be taken into consideration for women highly motivated to preserve sexual function so long as strict oncologic inclusion criteria are met [1]. In addition, though the oncological equivalence of open and robotic RC has been extensively assessed, [20–22] and the robotic approach has been widely adopted in the male sex-sparing counterpart with excellent functional results [23], there are no reports on sex-RARC in female patients.

Hence, in this paper and in the accompanying video, we firstly described surgical steps of sex-RARC in female patients and reported perioperative and functional outcomes of our initial series with a minimum 1-year follow-up. In the video, we clearly highlight the differences with the standard technique with special attention paid to the preservation of the utero-vaginal component of the IHP. We strongly believe that robotic technology offers undebatable advantages in meticulously following and dissecting the appropriate surgical planes since IHP fibers are usually embedded in dense connective tissue, and consequently are not always easy to preserve. Despite the small cohort (11 patients), the excellent continence results (daytime 90.9% and nighttime 86.4% at one year) and the encouraging rate of sexually active patients (72.7% at one year) reflect the proper respect of the crucial anatomical structures and reinforce the efficacy of sex-RC in properly selected women. In addition, the oncological effectiveness with an absence of any recurrence corroborates our results.

Another important point of strength regarding the reliability of our results is our adoption of standardized questionnaires to assess the quality of life, urinary symptoms, and sexual activity (i.e., EORTC-C30, BLM30, and FSFI), which are rarely used in most studies. Moreover, the minimal invasiveness of the robotic approach represents a further issue to consider when considering young sexually and socially active women.

Furthermore, our technique may avoid devastating complications such as vaginal dehiscence and evisceration which have been reported after minimally invasive radical cystectomy [24,25].

Nevertheless, the present paper is not devoid of limitations. The small sample size, the strict inclusion criteria, and the need for advanced robotic surgical skills are significant limitations to a wide reproducibility of these outcomes in daily practice. Finally, BCa recurrence usually occurs within two years of radical cystectomy. In this respect, the follow-up duration might be inadequate.

#### **5. Conclusions**

In selected populations, sex-RARC-iN can be offered to female patients motivated to preserve sexual function as an oncologically safe procedure, associated with favorable functional outcomes. The meticulous anatomical preservation of utero-vaginal components of IHP represents the cornerstone of a quick and effective recovery of physiological functions in terms of urinary continence and sexual activity. A proper comparison of outcomes with the conventional RARC-iN technique requires properly designed prospective randomized trials.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2077-0383/9/2/577/s1. Table S1. Health-related quality of life assessment (EORTC QLQ-C30 questionnaire). Table S2. Bladdercancer-specific quality of life EORTC QLQ-BLM30 questionnaire. Table S3. Female Sexual Function Index (FSFI) questionnaire. Table S4. Baseline and clinical characteristics of sex-sparing and standard RARC. Table S5. Perioperative, pathologic, and functional characteristics of sex-sparing and standard RARC. Figure S1. Kaplan–Meier analysis comparing day-time continence recovery probabilities between sex-sparing and standard RARC cohorts.

**Author Contributions:** Conceptualization, G.T, G.S.; methodology, G.T., R.M., G.S.; formal analysis, G.T.; data curation, R.M., S.F., M.F., C.L., U.A., A.B., S.G.; writing—original draft preparation, G.T., R.M.; writing—review and editing, G.T., G.S.; supervision, M.G., G.S. All authors have read and agreed to the published version of the manuscript.

**Conflicts of Interest:** The authors declare no conflicts of interest.

#### **References**

1. Witjes, J.A.; Lebret, T.; Compérat, E.; Cowan, N.C.; De Santis, M.; Bruins, H.M.; Hernández, V.; Espinos, E.L.; Dunn, J.; Rouanne, M.; et al. Updated 2016 EAU Guidelines on Muscle-invasive and Metastatic Bladder Cancer. *Eur. Urol.* **2017**, *71*, 462–475. [CrossRef]


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **Clear Cell Adenocarcinoma of the Urinary Bladder Is a Glycogen-Rich Tumor with Poorer Prognosis**

#### **Zhengqiu Zhou 1, Connor J. Kinslow 2, Peng Wang 3, Bin Huang 4, Simon K. Cheng 2,5, Israel Deutsch 2,5, Matthew S. Gentry 1,6 and Ramon C. Sun 6,7,\***


Received: 3 December 2019; Accepted: 30 December 2019; Published: 3 January 2020

**Abstract:** Clear cell adenocarcinoma (CCA) is a rare variant of urinary bladder carcinoma with a glycogen-rich phenotype and unknown prognosis. Using the National Cancer Institute's surveillance, epidemiology, and end results (SEER) program database, we documented recent trends in incidence, mortality, demographical characteristics, and survival on this rare subtype of urinary bladder cancer. The overall age-adjusted incidence and mortality of CCA was 0.087 (95% confidence interval (CI): 0.069–0.107) and 0.064 (95% CI: 0.049–0.081) respectively per million population. In comparison to non-CCAs, CCAs were more commonly associated with younger age (<60 years old, *p* = 0.005), female (*p* < 0.001), black ethnicity (*p* = 0.001), grade III (*p* < 0.001), and higher AJCC 6th staging (*p* < 0.001). In addition, CCA patients more frequently received complete cystectomy (*p* < 0.001) and beam radiation (*p* < 0.001) than non-CCA patients. Our study showed a poorer prognosis of CCAs compared to all other carcinomas of the urinary bladder (*p* < 0.001), accounted for by higher tumor staging of CCA cases. This study adds to the growing evidence that glycogen-rich cancers may have unique characteristics affecting tumor aggressiveness and patient prognosis. Additional mechanistic studies are needed to assess whether it's the excess glycogen that contributes to the higher stage at diagnosis.

**Keywords:** glycogen; clear-cell adenocarcinoma; urinary bladder; SEER program database

#### **1. Introduction**

Glycogen, a multibranched polymer of glucose, serves as our body's main form of carbohydrate storage [1]. In the past decade, glycogen has become well-established that, in addition to its role in maintaining metabolic homeostasis in normal cells, it also has a crucial role in promoting tumor growth, especially under adverse conditions [2]. Under hypoxic conditions, which are commonly encountered by tumors cells, expression of transcription factor HIF1α increases glycogen accumulation [3]. Cancer cells have been shown to mobilize this excess glycogen via a p38α mitogen-activated protein kinase pathway to fuel cellular proliferation and metastasis [4]. Glycogen has also been proposed to maintain

the Warburg effect in tumor cells, providing a mechanism for survival during nutrient deprivation [5]. Furthermore, glycogen's inability to metabolize glycogen through small molecule inhibitors was able to induce apoptosis or senescence in tumor cells [6,7]. Altogether, cancer cells utilize glycogen as a way to alter its metabolic programing in order to adapt to the adverse tumor microenvironment and maintain tumor growth.

Aberrant glycogen deposits have been identified in tumors from multiple origins, including cancers of the breast, kidney, uterus, lung, head and neck, bladder, ovary, skin, brain and colorectal tumors [8–12]. They are often identified as "clear cell" due to the transparent and ovoid appearance seen on histological staining. A poorer prognosis has been documented in clear cell carcinomas of the kidney [13], uterus [14], ovaries [15] and breast [16]. However, due to the rarity of some these tumors, the prognostic implications in other types of "clear cell" cancers remain unclear.

Clear cell adenocarcinoma of the urinary bladder (CCA) is a rare histological growth pattern first reported by Dow and Young in 1968 [17]. These tumors contain sheets of uniform ovoid cells with clear cytoplasm containing abundant glycogen [18,19]. Since there are no distinguishing symptoms of CCA, diagnosis is based on histopathological identification of these characteristics. Due to its rarity, information on the characteristics and prognosis of CCA have been limited to case reports, with less than 50 cases reported to date [19–21]. The largest existing literature review was performed by Lu et al., consisting of 38 case reports [21]. The review supported surgical resection as initial treatment for CCA and noted a possible increase in metastasis risk compared to urothelial carcinomas. However, the study determined that the prognosis of CCA was unclear as longer follow up periods were needed to more accurately assess survival characteristics [21]. No incidence and mortality data have been reported yet.

As the first large-scale study to date, we utilized the National Cancer Institute's surveillance, epidemiology, and end results (SEER) program database to conduct a retrospective assessment of incidence, mortality, demographics, and survival for CCA. Based on the previous literature that has shown a link between glycogen rich tumors and tumor aggressiveness [13–16], our study aimed to assess whether similar prognostic outcomes exist for CCAs. Using 91 cases of CCA and 205,106 cases of other urinary bladder cancers (non-CCA) obtained from the SEER Program database, we identified a poorer prognosis attributed to higher staging at time for diagnosis for CCAs. Our study contributes to the growing body of evidence revealing a possible link between glycogen and tumor aggressiveness.

#### **2. Experimental Section**

#### *2.1. Data Source*

The SEER Program is the National Cancer Institute's authoritative source of information on cancer incidence and survival capturing approximately 34.6% of the US population [22]. It is populated with high quality population-based data from national cancer registries. Vital status is updated annually and routinely undergoes quality-control checks.

#### *2.2. Sample Selection and Coding*

Age-adjusted incidence and mortality rates were calculated using the SEER\*Stat Software (Version 8.3.6, National Cancer Institute, Bethesda, MD, USA) using all 91 cases of malignant cases of CCA of the urinary bladder and 205, 106 cases of non-CCA from 2004 to 2015 from the SEER Program database [23,24]. Incidence and mortality were age-adjusted by standardizing to the 2000 United States Census population. All other data collection and analysis were conducted as described previously [16,25]. We obtained the November, 2015 submission [26] and November, 2017 submission [27] from the SEER Program database and merged all identified cases of malignant cancers of the bladder identified by International Classification of Diseases-O-3 (ICD-O-3) codes C67.0–C67.9 from January 2004 to December 2015. Carcinomas of the bladder were determined based on the adapted classification scheme for adolescents and young adults. Cases of clear cell adenocarcinoma were identified by ICD-O-3 code 8310.

The following variables were collected and coded: AYA site recode, primary site, ICD-O-3 histology, age at diagnosis, sex, race, grade, American Joint Commission on Cancer (AJCC) 6th Edition Staging, AJCC 6th Edition TNM system, survival months, vital status, bone metastasis at diagnosis, brain metastasis at diagnosis, liver metastasis at diagnosis, lung metastasis at diagnosis, surgery, and radiation. Cases of AJCC 6th stage 0a and 0is were merged and referred to as "stage 0". Ta, Tis were merged and referred to as "Ta/Tis". T1, T1a, T1b, T1 NOS were merged and collectively referred to as "T1". T2, T2a, T2b, T2 NOS were merged and collectively referred to as "T2". T3, T3a, T3b, T3c, T3 NOS were merged and collectively referred to as "T3". T4, T4a, T4b, T4 NOS were merged and collectively referred to as "T4". The surgery codes 10 (local tumor destruction), 20 (local tumor excision), and 30 (partial cystectomy) were merged and collectively referred to as "local procedure/ partial cystectomy". Surgical codes 50 (simple/total/complete cystectomy), 60 (complete cystectomy with reconstruction), and 70 (pelvic exenteration) were combined, and collectively referred to as "complete cystectomy". Surgical codes 80 (cystectomy, NOS) and 90 (surgery, NOS) were combined and collectively referred to as "surgery, NOS". Detailed SEER database surgery codes are available at (https://seer.cancer.gov/manuals/2018/appendixc.html). Cases diagnosed at autopsy or that could have 0 days of follow-up were excluded all analyses except for incidence and mortality calculations.

#### *2.3. Statistical Analysis*

All statistical analysis was carried out using the IBM SPSS Statistics software package (version 25, International Business Machines Corporation, Armonk, NY, USA). The significance of incidence and mortality trends were calculated using linear regression analysis. Differences in demographic and clinical characteristics between CCA and non-CCA were determined using the Pearson's chi-square test. Median survival times were determined using the Kaplan–Meier method, and the significance was determined using the log-rank test. Multivariable analyses of overall survival were conducted using the Cox proportional hazards ratios (HR) model. Corresponding HR and 95% confidence intervals (CI) were estimated from the model. Two-tailed *p*-values < 0.05 were considered statistically significant.

#### **3. Results**

#### *3.1. Incidence and Mortality of CCA*

To assess recent trends in the incidence and mortality of CCA, we queried all cases of CCA from 2004 to 2015 in the SEER Program database. Over this period, the age-adjusted the incidence of CCA was 0.087 individuals per 1,000,000 (Supplementary Table S1). Our analysis suggested a downward trend in incidence over this period—a shift from 0.062 per 1,000,000 in 2004 to 0.057 per 1,000,000 individuals in 2015 with an annual decrease rate of 0.003. However, this trend was non-significant (*p* = 0.178, Supplementary Figure S1A). We further assessed incidence separated by gender (Supplementary Table S1). The incidence of CCA among female and males were similar, with a slight female predominance—0.091 and 0.084 per 10,000,000 for females and males respectively from 2004 to 2015 (Supplementary Table S1).

The mortality rate from 2004–2015 was 0.064 individuals per 1,000,000 with an increasing trend of 0.002 per year. This trend was also non-significant (*p* = 0.477, Supplementary Figure S1B, Supplementary Table S1). When separated by gender, male with CCA had higher mortality rate of 0.074 compared to 0.058 in females per 1,000,000 individuals (Table S1).

#### *3.2. Demographics and Clinical Characteristics*

To compare demographical and clinical characteristics of CCA to non-CCA cancers of the urinary bladder, we utilized cases of malignant carcinomas of the urinary bladder from 2004, when AJCC 6th staging information became available, to 2015, the most recent data available at time of analysis. We obtained 205,197 cases of malignant urinary bladder carcinoma. Of these, 91 cases (0.04%) were identified as CCA. The median follow-up time was 19 months with 45 deaths in these CCA patients. Amongst 205,106 cases of non-CCA patients, the median follow-up time was 23 months, with 68,951 recorded deaths. The median age at diagnosis of CCA was 70 years old and median age at diagnosis was 72 years old in non-CCA patients.

The demographical and clinical characteristics of the patient population are summarized in Table 1. Our results showed that CCA patients were more likely to be younger age (<60 years of age; *p* = 0.005), female (*p* < 0.001) and black (*p* = 0.001) than non-CCA patients. The larger proportion of female patients is consistent with our incidence analysis. CCA patients also had higher grade (*p* < 0.001), higher AJCC 6th staging (*p* < 0.001) including TNM staging (*p* values for T, N, M stage were *p* < 0.001, *p* < 0.001 and *p* < 0.001, respectively). The primary site of tumor location was significantly different between CCA and non-CCA patients (*p* < 0.001); CCA patients were more likely to have tumors in the trigone of bladder, bladder neck and urachus, whereas non-CCA tumors appeared mostly in the lateral wall of bladder. As expected with more advanced tumor staging, CCA patients showed higher likelihood of brain (*p* < 0.001) and liver (*p* = 0.028) metastasis. However, very few cases with metastasis were available; only a single case was available for brain metastasis and two cases for liver metastasis. Furthermore, our data showed that non-CCA patients were more likely to receive fewer radical treatments such as local procedure or partial cystectomy, while more CCA patients received complete cystectomies (*p* < 0.001). The majority of non-CCA patients did not receive radiation, while a greater number of CCA patients received beam radiation (*p* < 0.001).




**Table 1.** *Cont.*

Bolded are statistically significant *p*-values when comparing between clear cell adenocarcinoma to other carcinomas of the urinary bladder. NA—not applicable. <sup>a</sup> Variable only available for cases diagnosed after 2010. <sup>b</sup> Variable only available for cases diagnosed before 2013.

#### *3.3. Survival*

The median survival for CCA patients was 34 months with 5- and 10-year survival rates of 41%, 30%, respectively. The median survival for non-CCA patients was 87 months, with corresponding 5- and 10-year survival rates of 61% and 44%, respectively (Figure 1, *p* < 0.001). Using multivariable analysis accounting for age, sex, race, AJCC 6th stage, tumor grade, surgery, and radiation treatment, survival for CCA patients was no longer significantly poorer than non-CCA patients (HR: 0.93; 95% CI: 0.69–1.255; *p* = 0.636, Supplementary Table S2 left half). However, when staging was removed from same multivariable analysis, CCA survival remained significantly shorter than non-CCA patients (HR: 1.435, 95% CI: 1.064–1.936, *p* = 0.018, Supplementary Table S2 right half). Therefore, the histological

subtype CCA is not an independent prognostic factor for survival, but instead, it is the more advanced staging in CCA patients accounts for the survival difference between CCA and non-CCA patients.

**Number at Risk.** 


**Figure 1.** Kaplan–Meier curve and risk table of clear cell adenocarcinoma in comparison to other carcinomas of the urinary bladder.

To further confirm our finding that the worse prognosis is attributable for the higher staging, we stratified our CCA cases according to AJCC 6th staging and compared survival in patients with non-muscle invasive (AJCC 6th stage 0 and I), muscle-invasive (AJCC 6th stage II and III) and metastatic (AJCC 6th stage IV) pathology. As suspected, when stratified by non-muscle invasive, muscle-invasive, and metastatic cases, the survival durations were no longer significantly different between CCA and non-CCA cases (Table 2, *p* = 0.654, *p* = 0.653, *p* = 0.091 respectively).


**Table 2.** Survival comparison between clear cell adenocarcinoma and other urinary bladder cancers stratified by stage.

Moreover, when surgical procedure was assessed in each subgroup of patients stratified by staging, a significant difference in the survival of muscle-invasive CCA patients was observed. Patients receiving total cystectomy showed significantly greater survival probability than those receiving local procedures or partial cystectomy (*p* = 0.028, Figure 2A). However, for metastatic cases, no survival difference was observed based on surgical treatment received (*p* = 0.269, Figure 2B). Survival

comparisons for non-muscle invasive cases were unable to be conducted due to the large number of censored events, i.e., patients that did not die during the follow-up period.

**Figure 2.** Kaplan–Meier curves and risk tables demonstrating survival for (**A**) muscle invasive cases of CCA defined by AJCC 6th stage II and III and (**B**) metastatic CCA cases defined by AJCC 6th stage IV.

#### **4. Discussion**

Using the SEER program database, we documented incidence, mortality, demographics, and survival on a rare subtype of urinary bladder cancer. We identified that CCAs were more commonly associated with younger age, higher grade, female gender, black ethnicity, and have a higher risk of brain and liver metastasis. Although it was not present in any of the cases reported in the SEER program database, bone metastasis in CCAs has been reported in several previously published case reports [28,29]. The most common location of CCA identified from our study was from trigone and bladder neck. This finding is consistent with previous reviews that also documented these as common tumor locations [21,28]. More importantly, our study showed a poorer prognosis of CCAs compared to all other carcinomas of the urinary bladder attributable to the higher tumor staging of the CCA cases. The poorer prognosis was irrespective of age, sex, race, grade, surgery and radiation treatment. In muscle invasive cases of CCA, type of surgical treatment was a significant factor in determining survival—There was improved survival when treated with complete cystectomies, which is consistent with standard of care for carcinomas of the urinary bladder [30].

The capability for glycogen to enhance tumor survival in adverse conditions may result in a faster invasion of CCA, hence, higher staging at diagnosis. Glycogen stores provide an excess glucose supply that can be utilized in the hypoxic conditions of tumor microenvironment [7]. The glycogen breakdown also generates nucleotides critical for cell proliferation such as NAPDH, an essential reducing agent, through the pentose phosphate pathway [7]. Furthermore, the glycogen shunt has been proposed to sustain the Warburg effect, a phenomenon that causes cells to use glucose in glycolysis instead of oxidative phosphorylation even in presence of plentiful oxygen in cells [31]. During periods of decreased glucose availability, the glycogen shunt sustains the production of glycolytic intermediates and ATP through the Warburg effect, hence maintaining tumor growth in nutrient deprived conditions [5].

Recently, the glycogen debranching enzyme amylo-α-1, 6-glucosidase, 4-α-glucanotransferase (AGL) was shown to have tumor suppressor functions in a model of urothelial bladder cancer [32]. Loss of AGL increased tumor growth in vitro and in xenografted tumors accompanied by an increase in abnormal glycogen structures (limit dextrin) and decrease in normal glycogen. The study also showed an increase in aerobic glycolysis and increased lactate, consistent with a shift towards the Warburg effect. Similar to our results, patients with reduced AGL expression was also associated with a decrease in overall survival, but was no longer predictive of survival when examined in a multivariate model that included age, sex, stage, and grade [32]. The similarities of our findings in CCA suggest that the manipulation of glycogen accumulations in urothelial bladder tumors may induce characteristics that mimic CCA.

While most urinary bladder cancers are male predominant [33], it was an interesting finding that CCA seemed to have a female predominance. The higher proportion of female patients supports a possible mullerian origin of CCA which has been previously proposed due to its association with endometriosis and histological resemblance to clear cell cancers of female genital tract [19,34]. Moreover, it is well known that females with urinary bladder cancers are generally diagnosed with more advanced disease and have poorer prognosis than males [33,35]. However, our findings suggested that it was CCA males instead who had higher mortality than females. Collectively, the gender disparity between CCA and other urinary bladder cancers suggest that CCA is an entity with differing characteristics to other urinary bladder cancers. More mechanistic and clinical studies are needed to improve our understanding of how gender and its associated factors relate to CCA pathology and prognosis.

At this time, no tailored therapy exists for CCA. Patients typically undergo some form of surgical resection such as transurethral resection, total cystectomy, partial cystectomy or radical surgery accompanied by chemotherapy and/or radiation [30]. Our study suggests that those with muscle invasive disease had survival benefit from total cystectomy rather than partial cystectomy, although prospective studies are needed to confirm these findings. Further understanding of cancer glycogen metabolism may help us with new avenues of tailored disease treatment. No information with regards to chemotherapy treatment was included in this manuscript due to a lack of reliable data in the SEER program database at this time.

#### **5. Conclusions**

As the first large-scale study to date, we assessed the incidence, mortality, demographical/clinical characteristics, and survival of CCA, a rare, glycogen-rich variant of urinary bladder cancer. We found a poorer prognosis of CCAs compared to all other carcinomas of the urinary bladder that was attributable to the higher staging of these tumors. However, the limitations of the study include the retrospective study design, small number of cases of interest (i.e., CCA) in comparison to control cases (i.e., non-CCA), and reliability of the SEER program database. Additional prospective clinical studies are needed to confirm these findings. Mechanistic studies that assess signaling pathways linking glycogen and rate of tumor growth would be beneficial for improving the understanding of the link between glycogen and poorer patient prognosis, and help to identify novel, targeted therapies for these glycogen-rich cancers.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2077-0383/9/1/138/s1, Table S1: Incidence and mortality of clear cell carcinoma of the urinary bladder from 2004–2015 per million population, Table S2: Multivariable analysis of survival for all urinary bladder patients, Figure S1: (**A**) Incidence of clear cell adenocarcinoma per million individuals in the US population. (**B**) Mortality of clear cell adenocarcinoma per million individuals in the US population.

**Author Contributions:** Z.Z. conducted the data analysis and drafted the manuscript. C.J.K. assisted with data analysis. P.W., S.K.C. and I.D. assisted in providing clinical insights for the manuscript. B.H. provided guidance on statistical analyses. M.S.G. and R.C.S. conceptualized the manuscript. All authors have read and agreed to the published version of the manuscript.

*J. Clin. Med.* **2020**, *9*, 138

**Funding:** This study was supported by the St Baldrick's Career Development Award (Scholar), University of Kentucky Center for Cancer and Metabolism P20 GM121327, American Cancer Society institutional research grant #16-182-28, funding from the University of Kentucky Markey Cancer Center P30CA177558. Z.Z. is supported by the NIH National Center for Advancing Translational Sciences (grant number: UL1TR001998).

**Acknowledgments:** Special thanks to Gentry lab and Vander Kooi lab for the numerous discussions and continuous support.

**Conflicts of Interest:** The authors declare no conflict of interest. Cheng reports personal fees and non-financial support from AbbVie and Sanofi. Gentry and Sun report personal fees and non-financial support from Maze Therapeutics. However, AbbVie, Sanofi and Maze Therapeutics had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References and Notes**


*J. Clin. Med.* **2020**, *9*, 138

35. Cohn, J.A.; Vekhter, B.; Lyttle, C.; Steinberg, G.D.; Large, M.C. Sex disparities in diagnosis of bladder cancer after initial presentation with hematuria: A nationwide claims-based investigation. *Cancer* **2014**, *120*, 555–561. [CrossRef]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
