Neurourology and Urodynamics
Risk of Malignancy After Augmentation Cystoplasty: A Systematic Review ^t,2,3 and Ve ronique Phe 2,3 Xavier Biardeau,1* Emmanuel Chartier-Kastler,2 Morgan Roupre 2
1 Department of Urology, Lille University Hospital, Lille Nord de France University, Lille, France Department of Urology, AP-HP, Pitie-Salpe^triere Academic Hospital, Sorbonne Universites, Paris, France 3 GRC-05, Institut Universitaire de Cancerologie (IUC), University Paris-6, Paris, France
Objectives: To systematically evaluate the evidence regarding the risk of malignancy after augmentation cystoplasty. Method: A systematic review search was performed through Medline and Embase databases using the following key words: ‘‘cancer,’’ ‘‘malignant neoplasm,’’ ‘‘cystoplasty,’’ and ‘‘bladder augmentation’’ until November 2014. An article was considered relevant to this review if it focused on malignant tumors occurring after augmentation cystoplasty performed for benign bladder disease. Results: After screening, 57 articles were included in the synthesis. The level of evidence was usually poor and results should be interpreted with caution. The follow-up time probability to develop a malignant tumor(s) after augmentation cystoplasty ranged from 0–5.5% and estimated incidence ranged from 0 to 272.3 per 100,000 patients/year. Adenocarcinoma was the commonest histological type (51.6%). Malignant lesions predominantly occurred at the entero-urinary anastomosis (50%). The mean latency period was 19 years and most malignant lesions were diagnosed more than 10 years after surgery (90%). Long-term surveillance by cystoscopy is still controversial because of its lack of efficiency. Non-invasive techniques have been proposed and need further evaluations. Tumors were often diagnosed at an advanced stage within surveillance protocols, because of urinary tract related symptoms (64.1%). The carcinogenesis pathway is still not clearly understood but several factors are involved. Conclusion: Augmentation cystoplasty is associated with a risk of malignancy. Studies regarding carcinogenesis and surveillance strategies should be considered to develop a more efficient follow-up protocol and allow early diagnosis. Neurourol. Urodynam. # 2015 Wiley Periodicals, Inc. Key words: bladder; carcinogenesis; cancer; histological type; latency period; malignant neoplasm; tumor location INTRODUCTION
Augmentation cystoplasty is the gold standard treatment for refractory neurogenic overactivity or low bladder compliance.1 This procedure aims to protect the upper urinary tract by restoring a low-pressure bladder and to improve quality of life by achieving continence among neurological patients. Different types of gastro-enteric segments have been used, but the ileum is the most common. Augmentation cystoplasty provides good long-term functional outcomes. Indeed, >90% of patients report a significantly improved quality of life. Nocturnal bladder continence is obtained in >90% of patients, and 91–100% report diurnal bladder continence.2 However, augmentation cystoplasty is a risk factor for cancer. Few data on this risk are available from the literature because of its unusual incidence and the long latency period between surgery and the occurrence of a malignant tumor. The purpose of this review was to systematically evaluate the evidence regarding the risk of malignancy after augmentation cystoplasty. METHOD Search Strategy
Two authors (XB,VP) performed a systematic review search in February 2014 through Medline and Embase databases. A freetext protocol was used, including the following key words: ‘‘cancer,’’ ‘‘malignant neoplasm,’’ ‘‘cystoplasty,’’ and ‘‘bladder augmentation.’’ No time limit was used because of the poor incidence and the long latency period reported in the literature. A language filter was used to restrict identification to English and French articles. Cited references from selected studies were #
2015 Wiley Periodicals, Inc.
also retrieved. For completeness, an updated search (XB,VP) was performed in December 2014. Study Selection
An article was considered relevant to this review if it focused on malignant tumors occurring after augmentation cystoplasty performed for benign bladder disease and reported or assessed one or more following items: incidence, latency period, tumor location, histological type, treatment, prognosis, followup strategy and mechanism, and risk factors of carcinogenesis. We accepted all study designs. Studies on humans or animals and reported in vitro experiments were assessed. Data Extraction and Analysis
Two authors (XB,VP) extracted and analyzed data from all selected studies. Incidence was evaluated from retrospective cohorts reporting malignant tumors after augmentation cystoplasty. We reported the follow-uptime probability to develop a malignant tumor after augmentation cystoplasty and calculated the estimated incidence whenever possible with Estimated Incidence ¼ (Percentage of malignant tumors/Median follow-up)*1000. Data related to the latency period, type of bowel segment used, malignant tumor location, histological
Correspondence to: Xavier Biardeau, Department of Urology, CHRU Lille, 1 rue Polonovski, H^ opital Claude Huriez, Lille 59000, France. E-mail: [email protected] Received 7 January 2015; Accepted 4 March 2015 Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/nau.22775
2
Biardeau et al.
type, circumstances of diagnosis, treatment, and prognosis were extracted and analyzed from all selected reported cases. Latency period was defined as the delay between surgery and malignant tumor diagnosis. Tumor location was defined as the site from which malignant lesion primarily arose. It was classified as ‘‘native bladder,’’ ‘‘bowel segment,’’ ‘‘enterourinary anastomosis,’’ or ‘‘not determined.’’ Histological type was extracted ensuring it corresponded to the 2004 World Health Organization classification of tumors of the urinary system.3 Prognosis was assessed thanks to the tumor stage at diagnosis according to the 2009 TNM classification of urinary bladder cancer.4 The mechanisms and risk factors of carcinogenesis, and follow-up strategies were assessed from human and animal studies, and reported in vitro experiments. RESULTS
Fifty-seven articles were included in the synthesis (Fig. 1).5–61 Eight articles assessed the risk of malignancy associated with augmentation cystoplasty from retrospective cohorts.5–12 Thirty-one articles reported cases of malignant tumors after augmentation cystoplasty in human subjects. 5,9–11,14–40 Moreover, 14 cases were not referenced in Medline or Embase databases or not published in English or French languages, but €lble.14 reported by Filmer and Spencer13 and Austen and Ka Overall 64 malignant lesions were reported in the literature. Sixteen articles investigated the mechanisms and risk factors of
carcinogenesis,7,10,14,27,41–52 13 focused on follow-up strategies after augmentation cystoplasty5,8–10,29,43,45,47,53–58 and one reviewed the risk of malignancy after augmentation cystoplasty for congenital abnormality.61 Incidence
For the last two decades, only a few authors used retrospective cohorts to evaluate the risk of malignancy after augmentation cystoplasty. The follow-up time probability to develop a malignant tumor after augmentation cystoplasty, using different types of bowel segments, ranged from 0 to 5.5% (Table I).5–12 The estimated incidence ranged from 0 to 272.3 per 100,000 patients/year5–12 while the worldwide age standardized incidence rate of urinary bladder cancer was recently estimated from 2.5 to 10.1 per 100,000 patients/year.62 These studies did not report a significant difference in the incidence of cancer between the three different types of bowel segment used. This has been supported in a retrospective study by Hamid et al.45 who evaluated the benefits of performing a yearly cystoscopy after augmentation cystoplasty in 92 patients. Although no malignant tumors were found after a mean time of 14 years following surgery, the authors reported chronic inflammatory changes and villous atrophy in similar proportions between colocystoplasty and ileocystoplasty procedures. The authors found no correlations between the biopsies and the segment of bowel used. Little et al.59 reported
Records identified through Medline and Embase databases searching (n=645)
Records after duplicates remove (n=615)
Records excluded (n=537): Not relevant to this review (n=537)
Records selected for full-text evaluation (n=78)
v Full-text articles excluded (n=21): Not relevant to this review (n=16) Not published in English or French (n=5)
Studies included in synthesis (n=57)
Fig. 1. Flow diagram underlying the search strategy and the study selection process for this review.
Neurourology and Urodynamics DOI 10.1002/nau
Risk of Malignancy After Augmentation Cystoplasty
3
TABLE I. Retrospective Evaluation for Malignant Lesions Prevalence After Augmentation Cystoplasty Patients Authors
n
Ali-El-Dein5 Husmann and Rathbun Higuchi et al.7
Type of cystoplasty
a
Kalble et al.8 Castellan et al.9 Vemulakonda et al.10 Soergel et al.11 Kispal et al.12 b
6 a
54 109 44 109 44 233 20 119 72 260 80
Ileocystoplasty Ileocystoplasty Colocystoplasty Ileocystoplasty Colocystoplasty Ileocystoplasty Colocystoplasty Gastrocystoplasty Gastrocystoplasty Various Various
Malignant tumors
Follow-up period (years)
Estimated incidence
n
%
Median (min–max)
/100,000 patients/year
3 5 2 5 2 2 0 4 2 3 0
5.5 4.6 4.6 4.6 4.6 0.9 0.0 3.4 2.8 1.2 0.0
20.2 (4–32) 28 (10–53) 28 (10–53) 27 (10–53) 27 (10–53) 10.5 (1–24) 4.9 (2–13) _ >10 >10 8.6 (1–20)
272.3 164.3 164.3 170.4 170.4 85.7 0 _ _ _ 0
a
Higuchi et al.7 and Husmann and Rathbun6 reported the same retrospective cohort in their two different studies. Kispal et al.12 performed regular systematic tissue samples from native bladder and gastrointestinal segments in 80 children after augmentation ileo-, colo-, or gastrocystoplasty. Three patients had pre-malignant histological changes diagnosed at 9,10, and 14 years after surgery.
b
similar results in a study investigating histopathological changes in rat bladders 21.5 months after augmentation cystoplasty using various gastrointestinal segments. The incidence of malignant tumors was not significantly different after gastro-, ileo- or colocystoplasty. Regarding the percentage of gastric, ileal or colonic segments generally used in augmentation cystoplasty, no data are available in the literature. However, even if the ideal bowel segment for enterocystoplasty remains controversial, all authors agree with ileum being currently the most used gut segment for bladder augmentation. Latency Period
The median latency period until a reported malignant tumor following augmentation cystoplasty was 19 years (range: 0.25–42) (Table II). Malignant neoplasms were diagnosed at 21 years (range: 2–42) after ileocystoplasty, at 20 years (range: 0.25–28) after colocystoplasty, and at 14 years (range: 10–15) after gastrocystoplasty. Ninety percent of tumors occured >10 years after surgery. Mean age at diagnosis was 44 years (range: 19–85) among the 58 patients whose age was reported while it is currently estimated to be about 75-years-old in the general population in both the USA and Western Europe.62
Among the 64 reported cases, 41 (64.1%) patients had urinary tract related symptoms reported at the time of diagnosis. Among them, 29 (70.7%) had a gross hematuria, 5 (12.2%) had an acute renal failure, 3 (7.3%) had recurrent urinary tract infections, 3 (7.3%) had symptomatic hydronephrosis with lumbar pain, and 1 (2.4%) had non-specific lower urinary tract symptoms. Treatments
The treatment was specified for all the selected reported cases. Radical surgery with radical cystectomy or pelvic exenteration was the most often performed procedure (68.7%) (Table V). Prognosis
The tumor stage at diagnosis was specified for 34 (53.1%) of the 64 reported cases. Among them, 12 (35.3%) patients had a locally advanced tumor found on histopathological analyses (10 pT3, 2 pT4). The initial lymph-node status was reported for 20 (31.2%) patients: among them, eight (40%) had lymphnode involvement. The initial metastasis status was specified for 27 (42.2%) patients: among them, 16 (59.3%) had distant metastasis at diagnosis.
Tumor Location
Most of the 64 malignant tumors after augmentation cystoplasty were located at the entero-urinary anastomosis (50%) (Table III). Histological Types
Adenocarcinoma was the commonest histological type after augmentation cystoplasty. Indeed, among the 64 reported cases, 51.6% were adenocarcinoma (Table IV). One case was reported as a leiomyosarcoma,14 a second as a carcinoid tumor,20 and two others as small cell carcinomas.16,30. The occurrence of these four tumors seemed to be coincidental and not related to augmentation cystoplasty. Circumstances of Diagnosis
Most of malignant tumors were diagnosed because of suspicious symptoms arising after augmentation cystoplasty. Neurourology and Urodynamics DOI 10.1002/nau
Follow-Up Cystoscopic biopsies. Most authors recommend performing an
annual cystoscopy after augmentation cystoplasty to detect tumors, starting at 10 years after surgery because of the long latency period between cystoplasty and the occurrence of malignant tumors.5,8,45,61 However, a lack of agreement on regular surveillance has been raised several times.9,10,47,53 Higuchi et al.53 evaluated the effectiveness of annual cystoscopy after following 55 patients for >10 years after surgery. They discontinued surveillance for the first 5 years after initiating the protocol because of the low rate and costs of the follow-up. Indeed, no malignant tumors were reported after 250 cystoscopies. Thus, for Higuchi et al.,54 the data strongly suggest that surveillance cystoscopy did not meet all the criteria needed for a successful screening protocol (Incidence should be clearly defined and frequent enough to make follow-up financially reasonable, the cancer must have a high prevalence at low stage and the screening test should have
4
Biardeau et al.
TABLE II. Age at Surgery, Age at Diagnosis, and Latency Period for Reported Malignant Lesions After Augmentation Cystoplasty Authors
Type of cystoplasty
Age at surgerya
Age at diagnosisa
Latency perioda
Ali-El-Dein et al.5 Ali-El-Dein et al.5 Ali-El-Dein et al.5 Austen and Kalble14 Austen and Kalble14 Barrington et al.15 Barrington et al.15 Barrington et al.15 Barrington et al.15 Bruyneel et al.16 Carr and Herschorn17 Egbertb El Otmany et al.18 Fernandez-Arjona et al.19 Hasegawab Jackson et al.20 Jung et al.21 Kamidonob Koizumi et al.c Lane and Shah22 Lane and Shah22 Leedhamb Louis et al.23 Moudouni et al.24 Nahas et al.25 Smith et al.b Soergel et al.11 Soergel et al.11 €ckle et al.c Sto Stone et al.26 Stone et al.26 Stone et al.26 Sung et al.27 Sung et al.27 Takasakib Ulmer et al.28 Bono Arino et al.c Docimo et al.29 Golomb et al.30 Golomb et al.30 Grainger et al.31 Gregoire et al.32 Harzmannb Husillos Alonso et al.33 Kadow et al.34 Llarena Ibarguren et al.c Shokeir et al.35 Soergel et al.11 Steg et al.b €ckle et al.c Sto Sung et al.27 Sung et al.27 Tellez Martinez-Fornes et al.c Balachandra et al.36 Baydar et al.37 Castellan et al.9 Castellan et al.9 Castellan et al.9 Castellan et al.9 Fernandez et al.38 Qiu et al.39 Vemulakonda et al.10 Vemulakonda et al.10 Zhang et al.40
Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocaecocystoplasty Ileocaecocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Colocystoplasty Caecocystoplasty Colocystoplasty Colocystoplasty Caecocystoplasty Caecocystoplasty Colocystoplasty Colocystoplasty Colocystoplasty Colocystoplasty Colocystoplasty Caecocystoplasty Colocystoplasty Colocystoplasty Caecocystoplasty Caecocystoplasty Colocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty
40 19 27 – – 42 23 28 18 37 26 24 29 47 13 73 26 27 18 72 72 36 38 31 20 26 24 8 – 49 31 18 24 8 21 19 22 23 32 45 69 64 27 43 39 – 23 24 37 – 23 11 – 54 22 13 22 6 18 13 59 6 5 5
53 43 58 – – 46 33 50 42 59 30 43 60 69 42 85 62 47 55 74 75 52 74 68 45 43 37 29 – 54 53 42 37 29 42 61 49 43 40 69 69 72 44 66 53 – 42 44 58 – 44 39 – 64 36 24 34 20 32 25 73 19 20 20
13 24 31 13.5 36.5 4 10 22 24 22 4 19 31 22 29 12 36 20 37 2 3 16 36 37 25 17 13 21 30 5 22 24 13 21 21 42 27 20 8 24 0.25 8 17 23 14 19 19 20 21 17 21 28 12 10 14 11 12 14 14 12 14 13 15 15
a
Age at surgery, age at diagnosis, and latency period expressed in years. Cases reported by Filmer and Spencer13 and Austen and K€ alble14. €lble14. Cases reported by Austen and Ka
b c
Neurourology and Urodynamics DOI 10.1002/nau
Risk of Malignancy After Augmentation Cystoplasty TABLE III. Tumor Location for Reported Malignant Lesions After Augmentation Cystoplasty
Ileocystoplasty Native Bladder Bowel segment Entero-urinary anastomosis Not determined Overall Colocystoplasty Native Bladder Bowel segment Entero-urinary anastomosis Not determined Overall Gastrocystoplasty Native Bladder Bowel segment Entero-urinary anastomosis Not determined Overall Overall Native Bladder Bowel segment Entero-urinary anastomosis Not determined Overall
n
%
12 4 19 1 36
33.3 11.1 52.8 2.8 100
2 3 9 3 17
11.8 17.6 53.0 17.6 100
0 3 4 4 11
0.0 27.2 36.4 36.4 100
14 10 32 8 64
21.9 15.6 50.0 12.5 100
a high specificity and sensitivity). Similarly, Kokorowski et al.55 concluded that annual cystoscopy and cytology were unlikely to be cost effective after augmentation cystoplasty. Thus, routine histological surveillance for malignancy or premalignant changes has been promoted by many authors. Some recommended a biopsy only if there were suspicious lesions. Others, have advocated systematic annual biopsy.43,56 The lack of prospective histological data has not allowed firm recommendations of when biopsy surveillance should be proposed. Urinary cytology. Limited information exists on the cytopathological features of urine from patients after augmentation cystoplasty. For Higuchi et al.,54 the presence of an enteric epithelium in a urinary reservoir greatly impacts on cytology effectiveness, which seems to be associated with poor sensitivity. Indeed, exfoliated enteric epithelium cells, chronic pyuria, and intermittent catheterization can cause cellular artifacts, such as increased nuclear-to-cytoplasmic ratio, nuclear atypia, and papillary aggregation. Buson et al.57 evaluated the efficiency of cytology to detect malignant tumors in 86 rats at 1 year after augmentation cystoplasty; The results were negative for neoplasic cells in all tumor cases. Fluorescent in-situ hybridization (FISH). Ivil et al.58 evaluated FISH of exfoliated bladder cells to predict tumor formation in 15 patients after augmentation cystoplasty. They reported that FISH could detect chromosomal losses and gains suggestive of genetic instability, and suggested that FISH could have a greater sensitivity than conventional urine cytology after augmentation cystoplasty.
TABLE IV. Histological Type for Reported Malignant Lesions After Augmentation Cystoplasty
Ileocystoplasty Adenocarcinoma Urothelial cell carcinoma Signet ring cell carcinoma Squamous cell carcinoma Small cell carcinoma Carcinoid tumor Leiomyosarcoma Overall Colocystoplasty Adenocarcinoma Urothelial cell carcinoma Signet ring cell carcinoma Squamous cell carcinoma Small cell carcinoma Carcinoid tumor Leiomyosarcoma Overall Gastrocystoplasty Adenocarcinoma Urothelial cell carcinoma Signet ring cell carcinoma Squamous cell carcinoma Small cell carcinoma Carcinoid tumor Leiomyosarcoma Unknown Overall Overall Adenocarcinoma Urothelial cell carcinoma Signet ring cell carcinoma Squamous cell carcinoma Small cell carcinoma Carcinoid tumor Leiomyosarcoma Unknown Overall
Neurourology and Urodynamics DOI 10.1002/nau
n
%
19 8 3 3 1 1 1 36
52.8 22.2 8.4 8.4 2.8 2.8 2.8 100
7 6 2 1 1 0 0 17
41.1 35.3 11.8 5.9 5.9 0.0 0.0 100
7 1 2 0 0 0 0 1 11
63.6 9.1 18.2 0.0 0.0 0.0 0.0 9.1 100
33 15 7 4 2 1 1 1 64
51.6 23.4 10.9 6.2 3.1 1.6 1.6 1.6 100
analysis detected clonal abnormalities in the urine and blood samples of one patient who developed an adenocarcinoma after augmentation cystoplasty. Upper urinary tract imaging. A few authors reported tumors diagnosed because of non symptomatic hydronephrosis and TABLE V. Treatments Performed for Reported Malignant Lesions After Augmentation Cystoplasty Treatment
n
%
Surgical treatmemt Radical surgery Conservative surgery
53 44 9 6 3 3 0 1 6 4 1 1 3
82.8 68.7 14.1 9.4 4.7 4.7 0.0 1.6 9.7 6.5 1.6 1.6 4.8
Adjuvant treatment
Polymerase chain reaction (PCR)-based microsatellite analysis.
PCR-based microsatellite analysis detects cancerspecific alterations in DNA structure. Preliminary data suggested that microsatellite analysis could detect 90–95% of bladder tumors. Moreover, it could detect DNAspecific alterations characteristic of either intestinal or bladder cancer. Docimo et al.29 reported that PCR-based microsatellite
5
Other treatment Chemotherapy Radiotherapy Radio-chemotherapy No specific treatment
Partial cystectomy Trans-urethral resection Adjuvant chemotherapy Adjuvant radiotherapy Adjuvant radio-chemotherapy
6
Biardeau et al.
recommended regular, annual, upper urinary-tract imaging, such as ultrasound imagery or computerized tomography.9,10 However, benefits and frequency of a regular surveillance imaging has never been assessed. Medical history and clinical findings. Some authors have recommended an alternative surveillance protocol based on patients’ medical history and clinical findings, to look for warning signs such as recurrent urinary-tract infections, bladder or pelvic pain, gross hematuria, persistant microhematuria. This approach could lead to suspicion and subsequent rapid investigations for a malignant tumor.45,53 Carcinogenesis
The exact mechanism of carcinogenesis after augmentation cystoplasty remains unknown and may be difficult to identify because of the long latency period between exposure and the occurrence of a malignant tumor. From our research, it appeared to be a complex multistaged and multifactorial process that could follow different pathways compared to conventional urothelial carcinoma or colonic adenocarcinoma. Sung et al.27 reported that malignant tumors associated with augmentation cystoplasty exhibited distinct morphological, immunohistochemical, and genetic characteristics compared to conventional urothelial carcinoma. They concluded that these malignant tumors were distinct variants rather than conventional urothelial carcinomas with mixed differentiation. Bacteriuria–N-nitro compounds. Bacteriuria was the first and main factor incriminated in carcinogenesis. Indeed, several authors initially reported a significant correlation between histological abnormalities (inflammation, metaplasia, dysplasia) and bacterial growth in urine cultures after different types of augmentation cystoplasty.41–43 Catalysation of urinary nitrate by bacteria in the augmented bladder has been reported to increase N-nitrosamine in infected or colonized urine.44 These N-nitro compounds were thought to act as initiators of carcinogenesis. However, this major role is now controversial as several recent studies have shown that malignant transformation and urinary-tract colonization or infection were not correlated.7,45,46,61 Chronic inflammation. A chronic inflammatory response has been implicated to release substances from inflammatory cells that could initiate carcinogenesis or sensitize the epithelium to carcinogens14 by increasing cell turnover.47 Basic fibroblast growth factor, inflammatory mediators, and substances, such as prostaglandin and other eicosanoids, cytokines, cyclooxygenase 2, and free oxygen radicals, have been implicated in the multistep carcinogenesis of different human cancers.14,48 Several in vitro and in vivo animal studies have reported an increased level of these substances in chronic inflammatory tissue and at the entero-urinary anastomosis.49 Chronic inflammation after bladder augmentation has been thought to be a consequence of bacteriuria, bowel mucus in contact with the urinary tract, urinary stasis in contact with the bowel segment, urinary pH changes, and stones. Most of these risk factors are theoretically reduced in gastrocystoplasty compared to enterocystoplasty because of gastric-acid secretion. However, the incidence of malignant tumor is not significantly different between these types of augmentation cystoplasty.10 This argues against the major role of bacterial colonization or infection, and of stones, mucus or urinary pH to cause carcinogenesis. Urinary hyper-osmolal conditions. Dixon et al.50 demonstrated that, in vitro, cell lines derived from gastric or colonic epithelium exposed to hyper-osmolal conditions showed an Neurourology and Urodynamics DOI 10.1002/nau
inadequate response to DNA damage. This was not observed with bladder-derived cells. Inherent abnormality within dysfunctional bladders. Higuchi et al.7 retrospectively compared a group of 145 patients who underwent an augmentation cystoplasty for bladder dysfunction with a second group of 145 patients with a similar bladder dysfunction and had no surgery. They reported no significant difference in the incidence of bladder cancer between the two groups based on the cause of bladder dysfunction or the type of bladder cancer. This underlines the inherent risk of malignancy in dysfunctional bladders whether an augmentation cystoplasty is performed or not. Perturbation in cell-to-cell interaction. A few authors suggested a key role of cell-to-cell interactions in carcinogenesis after augmentation cystoplasty.61 Gitlin et al.51 studied histological changes in the bladder of dogs at 4 months after gastrocystoplasty and ileocystoplasty. They reported the development of other hyperplasic transitional epithelial growths at the entero-vesical and gastro-vesical junctions. At the anastomosis, the cells expressed muco-substances, suggesting that they were undergoing glandular metaplasia. Li et al.52 suggested that the gastrointestinal bladder anastomosis constituted a conflicting zone of cell-to-cell interactions, in which the epithelium was receiving signals to both maintain the urothelial phenotype as well as to change intestinal phenotype. Their data demonstrated that the intestinal stroma was able to transform urothelium into glandular epithelium. Appanna et al.47 reported a number of genetic abnormalities in bladder segments from most patients after augmentation cystoplasty and found that urothelium adjacent to the anastomosis exhibited a greater degree of genetic instability than that distant to the anastomosis. Moreover, Hamid et al.45 reported that chronic inflammation and villous atrophy were more common in augmentation cystoplasty than in substitution cystoplasty, where the conflicting zone is smaller. These results argue for perturbation in the cell-to-cell interaction caused by the juxtaposition of these two organ systems.
DISCUSSION
The poor level of evidence of the included records prevents us from drawing any definitive conclusion and reported results should be interpreted with caution. However, this review clearly contributes to the assessment of the currently available evidence about the risk of malignancy after augmentation cystoplasty. Limitations Incidence. The estimation of incidence was usually based on retrospective data and often lacked statistical power due to the small numbers within cohorts and the short follow-up periods, Furthermore, we think that risk of malignancy after augmentation cystoplasty could be different in the future because of recent changes in indications. Indeed, this surgical procedure was initially used in the 1950s to treat small-retracted bladders secondary to urinary tuberculosis or schistosomiasis. However, it is now mostly used to treat refractory neurogenic overactivity or low bladder compliance. Latency period. For several authors, in case of a short latency period (
Risk of Malignancy After Augmentation Cystoplasty: A Systematic Review ^t,2,3 and Ve ronique Phe 2,3 Xavier Biardeau,1* Emmanuel Chartier-Kastler,2 Morgan Roupre 2
1 Department of Urology, Lille University Hospital, Lille Nord de France University, Lille, France Department of Urology, AP-HP, Pitie-Salpe^triere Academic Hospital, Sorbonne Universites, Paris, France 3 GRC-05, Institut Universitaire de Cancerologie (IUC), University Paris-6, Paris, France
Objectives: To systematically evaluate the evidence regarding the risk of malignancy after augmentation cystoplasty. Method: A systematic review search was performed through Medline and Embase databases using the following key words: ‘‘cancer,’’ ‘‘malignant neoplasm,’’ ‘‘cystoplasty,’’ and ‘‘bladder augmentation’’ until November 2014. An article was considered relevant to this review if it focused on malignant tumors occurring after augmentation cystoplasty performed for benign bladder disease. Results: After screening, 57 articles were included in the synthesis. The level of evidence was usually poor and results should be interpreted with caution. The follow-up time probability to develop a malignant tumor(s) after augmentation cystoplasty ranged from 0–5.5% and estimated incidence ranged from 0 to 272.3 per 100,000 patients/year. Adenocarcinoma was the commonest histological type (51.6%). Malignant lesions predominantly occurred at the entero-urinary anastomosis (50%). The mean latency period was 19 years and most malignant lesions were diagnosed more than 10 years after surgery (90%). Long-term surveillance by cystoscopy is still controversial because of its lack of efficiency. Non-invasive techniques have been proposed and need further evaluations. Tumors were often diagnosed at an advanced stage within surveillance protocols, because of urinary tract related symptoms (64.1%). The carcinogenesis pathway is still not clearly understood but several factors are involved. Conclusion: Augmentation cystoplasty is associated with a risk of malignancy. Studies regarding carcinogenesis and surveillance strategies should be considered to develop a more efficient follow-up protocol and allow early diagnosis. Neurourol. Urodynam. # 2015 Wiley Periodicals, Inc. Key words: bladder; carcinogenesis; cancer; histological type; latency period; malignant neoplasm; tumor location INTRODUCTION
Augmentation cystoplasty is the gold standard treatment for refractory neurogenic overactivity or low bladder compliance.1 This procedure aims to protect the upper urinary tract by restoring a low-pressure bladder and to improve quality of life by achieving continence among neurological patients. Different types of gastro-enteric segments have been used, but the ileum is the most common. Augmentation cystoplasty provides good long-term functional outcomes. Indeed, >90% of patients report a significantly improved quality of life. Nocturnal bladder continence is obtained in >90% of patients, and 91–100% report diurnal bladder continence.2 However, augmentation cystoplasty is a risk factor for cancer. Few data on this risk are available from the literature because of its unusual incidence and the long latency period between surgery and the occurrence of a malignant tumor. The purpose of this review was to systematically evaluate the evidence regarding the risk of malignancy after augmentation cystoplasty. METHOD Search Strategy
Two authors (XB,VP) performed a systematic review search in February 2014 through Medline and Embase databases. A freetext protocol was used, including the following key words: ‘‘cancer,’’ ‘‘malignant neoplasm,’’ ‘‘cystoplasty,’’ and ‘‘bladder augmentation.’’ No time limit was used because of the poor incidence and the long latency period reported in the literature. A language filter was used to restrict identification to English and French articles. Cited references from selected studies were #
2015 Wiley Periodicals, Inc.
also retrieved. For completeness, an updated search (XB,VP) was performed in December 2014. Study Selection
An article was considered relevant to this review if it focused on malignant tumors occurring after augmentation cystoplasty performed for benign bladder disease and reported or assessed one or more following items: incidence, latency period, tumor location, histological type, treatment, prognosis, followup strategy and mechanism, and risk factors of carcinogenesis. We accepted all study designs. Studies on humans or animals and reported in vitro experiments were assessed. Data Extraction and Analysis
Two authors (XB,VP) extracted and analyzed data from all selected studies. Incidence was evaluated from retrospective cohorts reporting malignant tumors after augmentation cystoplasty. We reported the follow-uptime probability to develop a malignant tumor after augmentation cystoplasty and calculated the estimated incidence whenever possible with Estimated Incidence ¼ (Percentage of malignant tumors/Median follow-up)*1000. Data related to the latency period, type of bowel segment used, malignant tumor location, histological
Correspondence to: Xavier Biardeau, Department of Urology, CHRU Lille, 1 rue Polonovski, H^ opital Claude Huriez, Lille 59000, France. E-mail: [email protected] Received 7 January 2015; Accepted 4 March 2015 Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/nau.22775
2
Biardeau et al.
type, circumstances of diagnosis, treatment, and prognosis were extracted and analyzed from all selected reported cases. Latency period was defined as the delay between surgery and malignant tumor diagnosis. Tumor location was defined as the site from which malignant lesion primarily arose. It was classified as ‘‘native bladder,’’ ‘‘bowel segment,’’ ‘‘enterourinary anastomosis,’’ or ‘‘not determined.’’ Histological type was extracted ensuring it corresponded to the 2004 World Health Organization classification of tumors of the urinary system.3 Prognosis was assessed thanks to the tumor stage at diagnosis according to the 2009 TNM classification of urinary bladder cancer.4 The mechanisms and risk factors of carcinogenesis, and follow-up strategies were assessed from human and animal studies, and reported in vitro experiments. RESULTS
Fifty-seven articles were included in the synthesis (Fig. 1).5–61 Eight articles assessed the risk of malignancy associated with augmentation cystoplasty from retrospective cohorts.5–12 Thirty-one articles reported cases of malignant tumors after augmentation cystoplasty in human subjects. 5,9–11,14–40 Moreover, 14 cases were not referenced in Medline or Embase databases or not published in English or French languages, but €lble.14 reported by Filmer and Spencer13 and Austen and Ka Overall 64 malignant lesions were reported in the literature. Sixteen articles investigated the mechanisms and risk factors of
carcinogenesis,7,10,14,27,41–52 13 focused on follow-up strategies after augmentation cystoplasty5,8–10,29,43,45,47,53–58 and one reviewed the risk of malignancy after augmentation cystoplasty for congenital abnormality.61 Incidence
For the last two decades, only a few authors used retrospective cohorts to evaluate the risk of malignancy after augmentation cystoplasty. The follow-up time probability to develop a malignant tumor after augmentation cystoplasty, using different types of bowel segments, ranged from 0 to 5.5% (Table I).5–12 The estimated incidence ranged from 0 to 272.3 per 100,000 patients/year5–12 while the worldwide age standardized incidence rate of urinary bladder cancer was recently estimated from 2.5 to 10.1 per 100,000 patients/year.62 These studies did not report a significant difference in the incidence of cancer between the three different types of bowel segment used. This has been supported in a retrospective study by Hamid et al.45 who evaluated the benefits of performing a yearly cystoscopy after augmentation cystoplasty in 92 patients. Although no malignant tumors were found after a mean time of 14 years following surgery, the authors reported chronic inflammatory changes and villous atrophy in similar proportions between colocystoplasty and ileocystoplasty procedures. The authors found no correlations between the biopsies and the segment of bowel used. Little et al.59 reported
Records identified through Medline and Embase databases searching (n=645)
Records after duplicates remove (n=615)
Records excluded (n=537): Not relevant to this review (n=537)
Records selected for full-text evaluation (n=78)
v Full-text articles excluded (n=21): Not relevant to this review (n=16) Not published in English or French (n=5)
Studies included in synthesis (n=57)
Fig. 1. Flow diagram underlying the search strategy and the study selection process for this review.
Neurourology and Urodynamics DOI 10.1002/nau
Risk of Malignancy After Augmentation Cystoplasty
3
TABLE I. Retrospective Evaluation for Malignant Lesions Prevalence After Augmentation Cystoplasty Patients Authors
n
Ali-El-Dein5 Husmann and Rathbun Higuchi et al.7
Type of cystoplasty
a
Kalble et al.8 Castellan et al.9 Vemulakonda et al.10 Soergel et al.11 Kispal et al.12 b
6 a
54 109 44 109 44 233 20 119 72 260 80
Ileocystoplasty Ileocystoplasty Colocystoplasty Ileocystoplasty Colocystoplasty Ileocystoplasty Colocystoplasty Gastrocystoplasty Gastrocystoplasty Various Various
Malignant tumors
Follow-up period (years)
Estimated incidence
n
%
Median (min–max)
/100,000 patients/year
3 5 2 5 2 2 0 4 2 3 0
5.5 4.6 4.6 4.6 4.6 0.9 0.0 3.4 2.8 1.2 0.0
20.2 (4–32) 28 (10–53) 28 (10–53) 27 (10–53) 27 (10–53) 10.5 (1–24) 4.9 (2–13) _ >10 >10 8.6 (1–20)
272.3 164.3 164.3 170.4 170.4 85.7 0 _ _ _ 0
a
Higuchi et al.7 and Husmann and Rathbun6 reported the same retrospective cohort in their two different studies. Kispal et al.12 performed regular systematic tissue samples from native bladder and gastrointestinal segments in 80 children after augmentation ileo-, colo-, or gastrocystoplasty. Three patients had pre-malignant histological changes diagnosed at 9,10, and 14 years after surgery.
b
similar results in a study investigating histopathological changes in rat bladders 21.5 months after augmentation cystoplasty using various gastrointestinal segments. The incidence of malignant tumors was not significantly different after gastro-, ileo- or colocystoplasty. Regarding the percentage of gastric, ileal or colonic segments generally used in augmentation cystoplasty, no data are available in the literature. However, even if the ideal bowel segment for enterocystoplasty remains controversial, all authors agree with ileum being currently the most used gut segment for bladder augmentation. Latency Period
The median latency period until a reported malignant tumor following augmentation cystoplasty was 19 years (range: 0.25–42) (Table II). Malignant neoplasms were diagnosed at 21 years (range: 2–42) after ileocystoplasty, at 20 years (range: 0.25–28) after colocystoplasty, and at 14 years (range: 10–15) after gastrocystoplasty. Ninety percent of tumors occured >10 years after surgery. Mean age at diagnosis was 44 years (range: 19–85) among the 58 patients whose age was reported while it is currently estimated to be about 75-years-old in the general population in both the USA and Western Europe.62
Among the 64 reported cases, 41 (64.1%) patients had urinary tract related symptoms reported at the time of diagnosis. Among them, 29 (70.7%) had a gross hematuria, 5 (12.2%) had an acute renal failure, 3 (7.3%) had recurrent urinary tract infections, 3 (7.3%) had symptomatic hydronephrosis with lumbar pain, and 1 (2.4%) had non-specific lower urinary tract symptoms. Treatments
The treatment was specified for all the selected reported cases. Radical surgery with radical cystectomy or pelvic exenteration was the most often performed procedure (68.7%) (Table V). Prognosis
The tumor stage at diagnosis was specified for 34 (53.1%) of the 64 reported cases. Among them, 12 (35.3%) patients had a locally advanced tumor found on histopathological analyses (10 pT3, 2 pT4). The initial lymph-node status was reported for 20 (31.2%) patients: among them, eight (40%) had lymphnode involvement. The initial metastasis status was specified for 27 (42.2%) patients: among them, 16 (59.3%) had distant metastasis at diagnosis.
Tumor Location
Most of the 64 malignant tumors after augmentation cystoplasty were located at the entero-urinary anastomosis (50%) (Table III). Histological Types
Adenocarcinoma was the commonest histological type after augmentation cystoplasty. Indeed, among the 64 reported cases, 51.6% were adenocarcinoma (Table IV). One case was reported as a leiomyosarcoma,14 a second as a carcinoid tumor,20 and two others as small cell carcinomas.16,30. The occurrence of these four tumors seemed to be coincidental and not related to augmentation cystoplasty. Circumstances of Diagnosis
Most of malignant tumors were diagnosed because of suspicious symptoms arising after augmentation cystoplasty. Neurourology and Urodynamics DOI 10.1002/nau
Follow-Up Cystoscopic biopsies. Most authors recommend performing an
annual cystoscopy after augmentation cystoplasty to detect tumors, starting at 10 years after surgery because of the long latency period between cystoplasty and the occurrence of malignant tumors.5,8,45,61 However, a lack of agreement on regular surveillance has been raised several times.9,10,47,53 Higuchi et al.53 evaluated the effectiveness of annual cystoscopy after following 55 patients for >10 years after surgery. They discontinued surveillance for the first 5 years after initiating the protocol because of the low rate and costs of the follow-up. Indeed, no malignant tumors were reported after 250 cystoscopies. Thus, for Higuchi et al.,54 the data strongly suggest that surveillance cystoscopy did not meet all the criteria needed for a successful screening protocol (Incidence should be clearly defined and frequent enough to make follow-up financially reasonable, the cancer must have a high prevalence at low stage and the screening test should have
4
Biardeau et al.
TABLE II. Age at Surgery, Age at Diagnosis, and Latency Period for Reported Malignant Lesions After Augmentation Cystoplasty Authors
Type of cystoplasty
Age at surgerya
Age at diagnosisa
Latency perioda
Ali-El-Dein et al.5 Ali-El-Dein et al.5 Ali-El-Dein et al.5 Austen and Kalble14 Austen and Kalble14 Barrington et al.15 Barrington et al.15 Barrington et al.15 Barrington et al.15 Bruyneel et al.16 Carr and Herschorn17 Egbertb El Otmany et al.18 Fernandez-Arjona et al.19 Hasegawab Jackson et al.20 Jung et al.21 Kamidonob Koizumi et al.c Lane and Shah22 Lane and Shah22 Leedhamb Louis et al.23 Moudouni et al.24 Nahas et al.25 Smith et al.b Soergel et al.11 Soergel et al.11 €ckle et al.c Sto Stone et al.26 Stone et al.26 Stone et al.26 Sung et al.27 Sung et al.27 Takasakib Ulmer et al.28 Bono Arino et al.c Docimo et al.29 Golomb et al.30 Golomb et al.30 Grainger et al.31 Gregoire et al.32 Harzmannb Husillos Alonso et al.33 Kadow et al.34 Llarena Ibarguren et al.c Shokeir et al.35 Soergel et al.11 Steg et al.b €ckle et al.c Sto Sung et al.27 Sung et al.27 Tellez Martinez-Fornes et al.c Balachandra et al.36 Baydar et al.37 Castellan et al.9 Castellan et al.9 Castellan et al.9 Castellan et al.9 Fernandez et al.38 Qiu et al.39 Vemulakonda et al.10 Vemulakonda et al.10 Zhang et al.40
Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocaecocystoplasty Ileocaecocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Ileocystoplasty Colocystoplasty Caecocystoplasty Colocystoplasty Colocystoplasty Caecocystoplasty Caecocystoplasty Colocystoplasty Colocystoplasty Colocystoplasty Colocystoplasty Colocystoplasty Caecocystoplasty Colocystoplasty Colocystoplasty Caecocystoplasty Caecocystoplasty Colocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty Gastrocystoplasty
40 19 27 – – 42 23 28 18 37 26 24 29 47 13 73 26 27 18 72 72 36 38 31 20 26 24 8 – 49 31 18 24 8 21 19 22 23 32 45 69 64 27 43 39 – 23 24 37 – 23 11 – 54 22 13 22 6 18 13 59 6 5 5
53 43 58 – – 46 33 50 42 59 30 43 60 69 42 85 62 47 55 74 75 52 74 68 45 43 37 29 – 54 53 42 37 29 42 61 49 43 40 69 69 72 44 66 53 – 42 44 58 – 44 39 – 64 36 24 34 20 32 25 73 19 20 20
13 24 31 13.5 36.5 4 10 22 24 22 4 19 31 22 29 12 36 20 37 2 3 16 36 37 25 17 13 21 30 5 22 24 13 21 21 42 27 20 8 24 0.25 8 17 23 14 19 19 20 21 17 21 28 12 10 14 11 12 14 14 12 14 13 15 15
a
Age at surgery, age at diagnosis, and latency period expressed in years. Cases reported by Filmer and Spencer13 and Austen and K€ alble14. €lble14. Cases reported by Austen and Ka
b c
Neurourology and Urodynamics DOI 10.1002/nau
Risk of Malignancy After Augmentation Cystoplasty TABLE III. Tumor Location for Reported Malignant Lesions After Augmentation Cystoplasty
Ileocystoplasty Native Bladder Bowel segment Entero-urinary anastomosis Not determined Overall Colocystoplasty Native Bladder Bowel segment Entero-urinary anastomosis Not determined Overall Gastrocystoplasty Native Bladder Bowel segment Entero-urinary anastomosis Not determined Overall Overall Native Bladder Bowel segment Entero-urinary anastomosis Not determined Overall
n
%
12 4 19 1 36
33.3 11.1 52.8 2.8 100
2 3 9 3 17
11.8 17.6 53.0 17.6 100
0 3 4 4 11
0.0 27.2 36.4 36.4 100
14 10 32 8 64
21.9 15.6 50.0 12.5 100
a high specificity and sensitivity). Similarly, Kokorowski et al.55 concluded that annual cystoscopy and cytology were unlikely to be cost effective after augmentation cystoplasty. Thus, routine histological surveillance for malignancy or premalignant changes has been promoted by many authors. Some recommended a biopsy only if there were suspicious lesions. Others, have advocated systematic annual biopsy.43,56 The lack of prospective histological data has not allowed firm recommendations of when biopsy surveillance should be proposed. Urinary cytology. Limited information exists on the cytopathological features of urine from patients after augmentation cystoplasty. For Higuchi et al.,54 the presence of an enteric epithelium in a urinary reservoir greatly impacts on cytology effectiveness, which seems to be associated with poor sensitivity. Indeed, exfoliated enteric epithelium cells, chronic pyuria, and intermittent catheterization can cause cellular artifacts, such as increased nuclear-to-cytoplasmic ratio, nuclear atypia, and papillary aggregation. Buson et al.57 evaluated the efficiency of cytology to detect malignant tumors in 86 rats at 1 year after augmentation cystoplasty; The results were negative for neoplasic cells in all tumor cases. Fluorescent in-situ hybridization (FISH). Ivil et al.58 evaluated FISH of exfoliated bladder cells to predict tumor formation in 15 patients after augmentation cystoplasty. They reported that FISH could detect chromosomal losses and gains suggestive of genetic instability, and suggested that FISH could have a greater sensitivity than conventional urine cytology after augmentation cystoplasty.
TABLE IV. Histological Type for Reported Malignant Lesions After Augmentation Cystoplasty
Ileocystoplasty Adenocarcinoma Urothelial cell carcinoma Signet ring cell carcinoma Squamous cell carcinoma Small cell carcinoma Carcinoid tumor Leiomyosarcoma Overall Colocystoplasty Adenocarcinoma Urothelial cell carcinoma Signet ring cell carcinoma Squamous cell carcinoma Small cell carcinoma Carcinoid tumor Leiomyosarcoma Overall Gastrocystoplasty Adenocarcinoma Urothelial cell carcinoma Signet ring cell carcinoma Squamous cell carcinoma Small cell carcinoma Carcinoid tumor Leiomyosarcoma Unknown Overall Overall Adenocarcinoma Urothelial cell carcinoma Signet ring cell carcinoma Squamous cell carcinoma Small cell carcinoma Carcinoid tumor Leiomyosarcoma Unknown Overall
Neurourology and Urodynamics DOI 10.1002/nau
n
%
19 8 3 3 1 1 1 36
52.8 22.2 8.4 8.4 2.8 2.8 2.8 100
7 6 2 1 1 0 0 17
41.1 35.3 11.8 5.9 5.9 0.0 0.0 100
7 1 2 0 0 0 0 1 11
63.6 9.1 18.2 0.0 0.0 0.0 0.0 9.1 100
33 15 7 4 2 1 1 1 64
51.6 23.4 10.9 6.2 3.1 1.6 1.6 1.6 100
analysis detected clonal abnormalities in the urine and blood samples of one patient who developed an adenocarcinoma after augmentation cystoplasty. Upper urinary tract imaging. A few authors reported tumors diagnosed because of non symptomatic hydronephrosis and TABLE V. Treatments Performed for Reported Malignant Lesions After Augmentation Cystoplasty Treatment
n
%
Surgical treatmemt Radical surgery Conservative surgery
53 44 9 6 3 3 0 1 6 4 1 1 3
82.8 68.7 14.1 9.4 4.7 4.7 0.0 1.6 9.7 6.5 1.6 1.6 4.8
Adjuvant treatment
Polymerase chain reaction (PCR)-based microsatellite analysis.
PCR-based microsatellite analysis detects cancerspecific alterations in DNA structure. Preliminary data suggested that microsatellite analysis could detect 90–95% of bladder tumors. Moreover, it could detect DNAspecific alterations characteristic of either intestinal or bladder cancer. Docimo et al.29 reported that PCR-based microsatellite
5
Other treatment Chemotherapy Radiotherapy Radio-chemotherapy No specific treatment
Partial cystectomy Trans-urethral resection Adjuvant chemotherapy Adjuvant radiotherapy Adjuvant radio-chemotherapy
6
Biardeau et al.
recommended regular, annual, upper urinary-tract imaging, such as ultrasound imagery or computerized tomography.9,10 However, benefits and frequency of a regular surveillance imaging has never been assessed. Medical history and clinical findings. Some authors have recommended an alternative surveillance protocol based on patients’ medical history and clinical findings, to look for warning signs such as recurrent urinary-tract infections, bladder or pelvic pain, gross hematuria, persistant microhematuria. This approach could lead to suspicion and subsequent rapid investigations for a malignant tumor.45,53 Carcinogenesis
The exact mechanism of carcinogenesis after augmentation cystoplasty remains unknown and may be difficult to identify because of the long latency period between exposure and the occurrence of a malignant tumor. From our research, it appeared to be a complex multistaged and multifactorial process that could follow different pathways compared to conventional urothelial carcinoma or colonic adenocarcinoma. Sung et al.27 reported that malignant tumors associated with augmentation cystoplasty exhibited distinct morphological, immunohistochemical, and genetic characteristics compared to conventional urothelial carcinoma. They concluded that these malignant tumors were distinct variants rather than conventional urothelial carcinomas with mixed differentiation. Bacteriuria–N-nitro compounds. Bacteriuria was the first and main factor incriminated in carcinogenesis. Indeed, several authors initially reported a significant correlation between histological abnormalities (inflammation, metaplasia, dysplasia) and bacterial growth in urine cultures after different types of augmentation cystoplasty.41–43 Catalysation of urinary nitrate by bacteria in the augmented bladder has been reported to increase N-nitrosamine in infected or colonized urine.44 These N-nitro compounds were thought to act as initiators of carcinogenesis. However, this major role is now controversial as several recent studies have shown that malignant transformation and urinary-tract colonization or infection were not correlated.7,45,46,61 Chronic inflammation. A chronic inflammatory response has been implicated to release substances from inflammatory cells that could initiate carcinogenesis or sensitize the epithelium to carcinogens14 by increasing cell turnover.47 Basic fibroblast growth factor, inflammatory mediators, and substances, such as prostaglandin and other eicosanoids, cytokines, cyclooxygenase 2, and free oxygen radicals, have been implicated in the multistep carcinogenesis of different human cancers.14,48 Several in vitro and in vivo animal studies have reported an increased level of these substances in chronic inflammatory tissue and at the entero-urinary anastomosis.49 Chronic inflammation after bladder augmentation has been thought to be a consequence of bacteriuria, bowel mucus in contact with the urinary tract, urinary stasis in contact with the bowel segment, urinary pH changes, and stones. Most of these risk factors are theoretically reduced in gastrocystoplasty compared to enterocystoplasty because of gastric-acid secretion. However, the incidence of malignant tumor is not significantly different between these types of augmentation cystoplasty.10 This argues against the major role of bacterial colonization or infection, and of stones, mucus or urinary pH to cause carcinogenesis. Urinary hyper-osmolal conditions. Dixon et al.50 demonstrated that, in vitro, cell lines derived from gastric or colonic epithelium exposed to hyper-osmolal conditions showed an Neurourology and Urodynamics DOI 10.1002/nau
inadequate response to DNA damage. This was not observed with bladder-derived cells. Inherent abnormality within dysfunctional bladders. Higuchi et al.7 retrospectively compared a group of 145 patients who underwent an augmentation cystoplasty for bladder dysfunction with a second group of 145 patients with a similar bladder dysfunction and had no surgery. They reported no significant difference in the incidence of bladder cancer between the two groups based on the cause of bladder dysfunction or the type of bladder cancer. This underlines the inherent risk of malignancy in dysfunctional bladders whether an augmentation cystoplasty is performed or not. Perturbation in cell-to-cell interaction. A few authors suggested a key role of cell-to-cell interactions in carcinogenesis after augmentation cystoplasty.61 Gitlin et al.51 studied histological changes in the bladder of dogs at 4 months after gastrocystoplasty and ileocystoplasty. They reported the development of other hyperplasic transitional epithelial growths at the entero-vesical and gastro-vesical junctions. At the anastomosis, the cells expressed muco-substances, suggesting that they were undergoing glandular metaplasia. Li et al.52 suggested that the gastrointestinal bladder anastomosis constituted a conflicting zone of cell-to-cell interactions, in which the epithelium was receiving signals to both maintain the urothelial phenotype as well as to change intestinal phenotype. Their data demonstrated that the intestinal stroma was able to transform urothelium into glandular epithelium. Appanna et al.47 reported a number of genetic abnormalities in bladder segments from most patients after augmentation cystoplasty and found that urothelium adjacent to the anastomosis exhibited a greater degree of genetic instability than that distant to the anastomosis. Moreover, Hamid et al.45 reported that chronic inflammation and villous atrophy were more common in augmentation cystoplasty than in substitution cystoplasty, where the conflicting zone is smaller. These results argue for perturbation in the cell-to-cell interaction caused by the juxtaposition of these two organ systems.
DISCUSSION
The poor level of evidence of the included records prevents us from drawing any definitive conclusion and reported results should be interpreted with caution. However, this review clearly contributes to the assessment of the currently available evidence about the risk of malignancy after augmentation cystoplasty. Limitations Incidence. The estimation of incidence was usually based on retrospective data and often lacked statistical power due to the small numbers within cohorts and the short follow-up periods, Furthermore, we think that risk of malignancy after augmentation cystoplasty could be different in the future because of recent changes in indications. Indeed, this surgical procedure was initially used in the 1950s to treat small-retracted bladders secondary to urinary tuberculosis or schistosomiasis. However, it is now mostly used to treat refractory neurogenic overactivity or low bladder compliance. Latency period. For several authors, in case of a short latency period (
