Environmental factors involved in carcinogenesis of urothelial cell carcinomas of the upper urinary tract
Morgan Rouprêt, Hôpital Pitié-Salpétrière, 47–83 Boulevard de l’hôpital, 75013 Paris, France.
Primary cancers of the ureter and renal pelvis are rare tumours, >90% of which are transitional cell carcinomas. Only ≈5% of urothelial tumours arise in the upper urinary tract (UUT). Many environmental factors contribute to the development of these cancers. Some are similar to bladder cancer-associated factors (tobacco, occupational exposure), while others are more specific to carcinogenesis of the UUT (phenacetine, Balkan endemic nephropathy [BEN], Chinese herb nephropathy or association with Blackfoot disease [BFD]). This review discusses the environmental factors involved in UUT carcinoma. Tobacco and occupational exposure remain the principal exogenous risk factors for developing these tumours. Conversely, carcinogenesis of UUT tumours resulting from phenacetine consumption has almost disappeared. Although the incidence of BEN is also on the decline, roles for aristolochic acid and the consumption of Chinese herbs in the physiopathology and induction of this nephropathy, respectively, have proposed. In Taiwan, the association of this tumour type with BFD and arsenic exposure remains unclear to date. As some genetic polymorphisms are associated with an increased risk of cancer or faster disease progression, there is variability in interindividual susceptibility to the development of UUT carcinoma when exposed to the aforementioned risk factors Cytosolic sulfotransferases (SULTs) catalyse the detoxification of many environmental chemicals but also in the bioactivation of dietary and other mutagens. Polymorphism of the SULT gene, is thought to confer susceptibility to upper tract tumours.
(upper urinary tract) urothelial cell carcinoma
Balkan endemic nephropathy
urothelial carcinoma genesis
Primary upper urinary tract urothelial cell carcinomas (UUT-UCCs) are rare tumours with an estimated incidence of 1–4 cases per 100 000 individuals per year. They represent only 5–6% of all UCCs . UUT-UCC distribution is restricted to the renal pelvis in 58% of cases, the ureter in 35% (73% of which are located in the distal ureter), both kidney and the ureter in 7% and a bilateral involvement in 2.5%. On the other hand, bladder cancer is ranked the fourth most common cancer type in men and the ninth in women worldwide, and more than half of all cases occur in developed countries . Pyelocaliceal tumours are about twice as common as ureteric tumours. In 8–13% of cases, a synchronous bladder cancer is present. Recurrence in the bladder or the contralateral urinary tract occurs in 30–51% of UUT-UCCs. The natural history of UUT-UCCs differs from that of bladder cancer: 60% of UUT-UCCs are invasive at diagnosis compared with only 15% of bladder tumours. For those with UCC of the bladder, the risk of UUT recurrence may be as high as 25%[1,4]. The mean age of occurrence is 65 years in most series, although a recent USA records study showed a trend towards diagnosis at an older age (70 years) . UUT-UCCs occur more frequently in men, with a male-to-female ratio of 3:1, but this discrepancy can probably be attributed to the greater likelihood of men being exposed to the effects of major carcinogens for many years . Although the mechanisms of carcinogenesis were thought to be similar throughout the urinary tract, recent epidemiological data and genetic studies suggest otherwise. It is now obvious that strong differences exist for tumour location and behaviour between the UUT and the lower urinary tracts . To date, many support the hypothesis that genetic susceptibility to cancer from genetic polymorphisms is enhanced by exposure to exogenous factors that induce specific gene–environment interactions and promote the development of UUT-UCC . In this short review, we present the environmental factors involved in UUT-UCC genesis.
The association of urothelial carcinoma genesis (UCG) with smoking exposure is complex and linked to multiple inhaled toxic substances (aromatic amine whith arylamine, benzopyrene, dimethylbenzanthracene). The aromatic amines are metabolized by the body into n-hydroxyalanine, which has a real carcinogenic activity. The detoxification of this derivative is modulated by several different enzyme systems (cytochrome P450s including CYP1A1, the glutathione S-transferases and n-acetyl transferases), any of which could harbour genetic polymorphisms that might explain individual susceptibility to developing UUT-UCCs . Exposure to tobacco increases the relative risk (RR) of developing a UUT-UCC to 2.5–7 . This risk is modulated by the number of years of exposure and by the number of cigarettes smoked every day. McLaughlin et al.  found that the estimated risk (odds ratio, OR) of developing UUT-UCCs ranged from 2.4 with consumption of <20 cigarettes per day to 4.8 with consumption of >40 cigarettes per day. The risk decreases by about 60–70% with an interruption of smoking for >10 years.
UUT-UCC ‘amino-tumours’, as described by Mazeman  in 1972, are related to occupational exposure to certain aromatic amines. These aromatic hydrocarbons have been and are used in many industries (dyes, textiles, rubber, chemicals, petrochemicals, coal) and research. Aromatic amines are responsible for the carcinogenicity of certain chemicals, including benzidine and β-naphtylanine [1,11]. These two pathogens have been phased out since 1960 and were eventually banned in most industrialized countries. In most cases, UUT-UCCs are secondary to a bladder amino-tumour. The average duration of exposure needed to develop a UUT-UCC is ≈7 years, with a latency period after the end of exposure of ≈20 years . The estimated risk (OR) of developing TCC when exposed to aromatic amines is 8.3 according to Shinka et al. .
Polycyclic aromatic hydrocarbons and nitrosamines have also been implicated in UCG. Clavel et al.  have shown that exposure to polycyclic aromatic hydrocarbons, adjusted for tobacco consumption and exposure to carcinogenic aromatic amines, led to an increased risk (OR) of developing TCC of 1.3, with a significant dose effect.
A role for chlorinated solvents (trichlorethylene, tetrachlorethylene) in UCG has also been suggested. These chemicals are usually used in metallurgy and printing. For exposure of this type, the estimated risk (OR) calculated by Pesch et al.  was 1.8. It is difficult to determine the share of each exposure among workers in these industries.
For 40 years, the regular consumption of phenacetin, used in various analgesic preparations, has been recognized as a risk factor for UUT-UCC. The nephrotoxicity of phenacetin was suspected for the first time in Sweden in 1961, when the first cases of UUT-UCCs related to consumption were reported .
There are two opposing hypotheses to explain phenacetin-associated carcinogenesis:
(i) Similar to aromatic amines, phenacetin has a direct mutagenic effect.
(ii) Phenacetin indirectly causes carcinogenesis by inducing nephrotoxicity through papillary necrosis. This could promote carcinogenesis or act as a cofactor in the presence of chronic irritation, infection or smoking.
There is now experimental and epidemiological evidence in support of the second hypothesis . The use of phenacetin was progressively abandoned in the 1980s and replaced by paracetamol in association with other analgesics. As the average latency for appearance of UUT-UCCs was ≈22 years, the number of cases described today is decreasing .
Specific epidemiological characteristics of this association include:
(i) RR of occurrence correlated with consumption of analgesics;
(ii) association with specific nephropathy may lead to end-stage renal failure (8–10% of patients with this nephropathy developed a UUT-UCC);
(iii) females reported that they consumed analgesics in response to greater levels of pain as compared with men;
(iv) tendency to bilateral lesions;
(v) no proven correlation between histologically high-grade tumour stages and abuse of phenacetin.
Analgesic compounds that lack phenacetin are not currently recognized as factors influencing the genesis of UUT-UCC.
BALKAN ENDEMIC NEPHROPATHY (BEN) AND CHINESE HERBS NEPHROPATHY
Since 1950, a dramatically high rate of UUT-UCC incidence has been described in some rural areas of the Balkans (60–100 times higher than the rest of the world). However, there has been a reduction in this rate over the last 20 years (only 11-times higher in the endemic area in 1998 compared with 57-times higher in 1988, according to Markovic et al. ). UUT-UCCs developed in these regions are associated with an endemic nephropathy called BEN [17–19]. BEN has been described as a proximal tubular dysfunction responsible for the presence of a low molecular weight protein and a dense interstitial fibrosis within the glomeruli.
Tumours in this geographical region have special characteristics:
- • increased occurrence of bilateral cases (8–10%)
- • lack of male predominance (some studies even found female-biased trends)
- • occurrence in rural areas
- • diagnosis of BEN ≈10 years before
Multiple environmental hypotheses have been advanced over the last 50 years to explain this nephropathy and its association with UUT-UCC (e.g. exposure to heavy metals, arsenic, nitrogen derivatives, viruses, bacteria, and mycotoxins as well as deficiency in selenium, calcium and magnesium). A role for mycotoxin ochratoxin A (OTA) has been widely investigated for 30 years [17–19]. However, the carcinogenic effects of OTA have only been shown in animals and induced histological lesions did not have the same characteristics as those found in BEN.
This association has been neglected lately in the light of recent publications on another nephropathy: the ‘Chinese herbs nephropathy’[17,20,21]. Between 1992 and 1993, 43 patients were hospitalized in Belgium for end-stage renal failure after ingestion of Chinese herbal products. Investigations revealed an error in the manufacture of the medicinal mixture with replacement of Stephania tetracta (‘Han Fang-ji’) by Aristolochia fangchi (‘Guang Fang-ji’). Almost 46% of these patients developed a UUT-UCC with histological and genetic features similar to those described for BEN. Several studies have shown the carcinogenic potential of the aristolochic acid contained in Aristolochia fangchi and Aristolochia clematis (a plant endemic to the Balkans). The aristolochic acid derivative d-aristoloctam causes a specific mutation of the p53 gene at codon 139 (A:T→T:A). This mutation is very rare in the unexposed population and is predominant in patients with Chinese herbs nephropathy or BEN who present UUT-UCC [20–22]. A possible role for A. clematis in BEN was previously mentioned, noting that the plant, which grows in the wheat fields of this area, likely contaminated bread flour [19–21]. This hypothesis, abandoned for 25 years, has recently been re-introduced [19–21]. The interindividual variability in developing UUT-UCC might be explained by different levels of activity of enzymes that metabolize toxic derivatives of aristolochic acid. These enzymes are modulated by tobacco exposure, environmental chemicals, drugs and familial genetic polymorphisms . Debelle et al.  recently proposed combining BEN and Chinese herbs nephropathy under one name: aristolochic acid nephropathy.
BLACKFOOT DISEASE (BFD)
A high incidence of UUT-UCC has also been described in Taiwan [24,25]. The population of the south-west coast of the island is particularly susceptible to UUT tumours where they represent 20–25% of UCCs. This geographical area corresponds to the endemic location of BFD, which is known to be a vasculitis. This vasculitis, caused by chronic exposure to arsenic pollution of water in artesian wells, has been the subject of numerous epidemiological studies over the past 45 years [24,25].
The characteristics of UUT-UCC in the endemic area of BFD are as follows [24,25]:
- • male : female ratio of 1:2
- • age of onset younger than usual (55–60 years)
- • ureteric tumours twice as common as the renal pelvis site.
However, the relationship between BFD and UUT-UCC remains unclear . There is also a high incidence of these tumours in the north-east end of the island (10–15% of UCC) where BFD is not a problem. In addition, most patients with UUT-UCC in these endemic areas have never been personally affected by BFD. Finally, as polluted wells are used equally by both men and women, the skewed sex ratio in these endemic areas cannot be explained solely by the toxic exposure.
The association between exposure to arsenic and UUT-UCC has never been formally proven. A high concentration of arsenic in water appears to be correlated with the development of bladder UCC, but cannot be the only relevant risk factor in this region . Arsenic exposure may simply be an isolated mutagenic agent in these endemic areas that, combined with other not-yet discovered factors spread across the island (i.e. other pollutants or even Chinese herbs) predisposes individuals to develop UUT-UCC [17,24,26].
There is a history of lithiasis in 5–8% of patients with UUT-UCC. However, the tumour incidence in lithiasis cases remains quite low at ≈1%. The chronic irritation or inflammation of the urothelium induced by the presence of the concretions as well as stasis from obstruction may promote tumour proliferation. Almost 60% of tumours developed in this region are not papillary epithelial tumours (squamous cell carcinoma). The RR of developing UUT-UCC in cases with a history of lithiasis varies from 1.2 to 2.5 [27,28]. This risk increases with female gender, renal pelvis calcifications, history of concomitant infections and the number of hospital-related disease lithiasis.
Chronic UTIs facilitate carcinogenesis by weakening the urothelium. The estimated risk (OR) to develop a UUT-UCC is 1.5–2 in cases of chronic UTI .
Cyclophosphamide and ifosfamide
Alkylating chemotherapy treatments induce UCC via the associated metabolites (acrolein). Chronic exposure to these molecules leads to the appearance of UUT-UCC (epidermoid pathology). The RR of developing UCC in this context would be 3.2 .
These products and particularly anthranoides and chemical laxatives have recently been implicated in UCG . Pommer et al.  have reported a RR of 9.62 for developing a UUT-UCC after using laxatives for a year. Evidence for the carcinogenic effect of laxatives were obtained only in animal studies . Their mechanism of action has not been explicated in humans. However, it seems that habitual intake of specific laxatives influences the development of UUT-UCCs .
Radiation-induced UCCs were seen in patients treated with radiotherapy for gynaecological cancers. However, the RR of UCC after radiotherapy (1.9) is still lower than that found in patients treated with chemotherapy. There is no known data that specifically addresses the risk of developing UUT-UCC .
OTHER RISK FACTORS
A role for hypertension has been mentioned, but the carcinogenic mechanisms remain obscure. The estimated risk (OR) for UUT-UCC is 1.3 in patients with hypertension .
Coffee, tea and yerba mate
Several studies have questioned whether the regular consumption of coffee or tea is a risk factor for developing UUT-UCC. However, no significant associations have been reported to date . In South America, consumption of yerba mate seems more directly responsible for UCG, as high levels of polycyclic aromatic hydrocarbons can result even when drinking responsibly. The RR of developing UCC from yerba mate consumption is estimated at 2.2, but no study to date has looked specifically at UUT-UCC .
Some genetic polymorphisms are associated with an increased risk of cancer or faster disease progression. Susceptibility means an increase in risk conferred by one or more polymorphisms in a given gene or genes, which expose the individual, family or group of individuals (ethnic/geographical variations) to the genotoxic effects of environmental carcinogens or promote tumour progression by determining a particular hormone or immune status. Differences in the ability to repress carcinogens may contribute to host susceptibility and be involved in risk of developing urothelial carcinomas. It will be important to determine the relative impacts of particular allelic variants of genes coding for enzymes involved in the metabolism of carcinogens and the molecular alterations in UUT-UCC. Only one polymorphism specific to UUT-UCC has been reported to date . The variant allele SULT1A1*2, which reduces sulfotransferase activity, enhances the risk of developing UUT-UCCs. Sulphation is an important step in the detoxification of many environmental chemicals but also in the bioactivation of dietary mutagens. Cytosolic sulfotransferases (SULTs) catalyse the conjugation of a sulphate group to a substrate. The two major gene families are the phenol (SULT1A1) and hydroxysteroid (SULT2A) sulfotransferases. SULT1A1 is expressed in several tissues, including liver, lung and kidney. The SULT1A1*2 polymorphism, which gives rise to an Arg213His substitution, is thought to confer susceptibility to various types of cancer. We have shown that His213 allele frequency is associated with a significantly higher risk of UUT-TCC. The association between the SULT1A1 polymorphism and bladder cancer is different, as the reduced enzymatic activity seems to actually protect against this tumour .
Risk factors for developing UUT-UCC are summarized in Table 1[9–17,25,27–31].
Table 1. Risk factors for developing UUT-UCC identified in published reports
|Tobacco||2.5–7 ||4–11 ||–|
|Aromatic amines||–||8.3 (ELUCC) ||–|
|Polycyclic aromatic hydrocarbons||–||1.3–1.6 (ELUCC) [12,13]||–|
|Chlorinated solvents||–||1.8 (ELUCC) ||–|
|Phenacetine||1.4–5.4 [14,15]||5.3–6.5 ||–|
|BEN||–||–||29.2/100 000 in endemic areas in 1998 |
|Chinese herbs nephropathy||–||–||40–46% of exposed patients in Europe, unknown in Asia |
|BFD||–||–||20–26% of ELUCC in endemic areas |
|Chronic infections||–||1.5–2 ||–|
|Cyclophosphamide||3.2 (ELUCC) ||–||–|
|External radiotherapy||1.9 (ELUCC) ||–||–|
|Arterial hypertension||–||1.3 ||–|
|Yerba mate||–||2.2 (ELUCC) ||–|
UUT-UCC can be distinguished from tumours of the bladder by certain molecular and genetic characteristics. In addition, particular environmental factors are specifically associated with UUT-UCC (phenacetin, aristolochic acid nephropathy, BFD) Our understanding of the carcinogenic mechanisms involved in the catabolism of these toxic chemicals is still quite limited. However, it is increasingly clear that interactions between genetic susceptibility and exposure to these environmental factors are responsible for the genesis of most sporadic UUT-UCCs.
CONFLICT OF INTEREST