SEARCH

SEARCH BY CITATION

Keywords:

  • urothelial carcinoma;
  • renal pelvis;
  • ureter;
  • sporadic;
  • hereditary

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

Study Type – Diagnostic (exploratory cohort)

Level of Evidence 3a

What's known on the subject? and What does the study add?

Hereditary non-polyposis colorectal cancer (HNPCC), also known as Lynch syndrome, is an autosomal dominant multi-organ cancer syndrome. Upper urinary tract urothelial carcinomas belong to HNPCC-related tumours and rank third within this group after colorectal and endometrial cancer. However, many urologists are not aware of this association and it is presumed that some hereditary cancers are misclassified as sporadic and that their incidence is underestimated. Consequently, family members of patients with upper urinary tract urothelial carcinomas secondary to HNPCC may be denied appropriate surveillance and early detection.

A significant proportion of patients (21.3%) with newly diagnosed upper urinary tract urothelial carcinomas may have underlying HNPCC. Demographic and epidemiological characteristics suggest different mechanisms of carcinogenesis among this population. Recognition of such potential is essential for appropriate clinical and genetic management of patients and family. In order to help to identify these patients, we propose a patient-specific checklist.

OBJECTIVE

  • • 
    To identify, based on previously described clinical criteria, hereditary upper urinary tract urothelial carcinomas (UUT-UCs) that are likely to be misclassified as sporadic although they may belong to the spectrum of hereditary non-polyposis colorectal cancer (HNPCC) associated cancers.

PATIENTS AND METHODS

  • • 
    We identified, using established clinical criteria, suspected hereditary UUT-UC among 1122 patients included in the French national database for UUT-UC.
  • • 
    Patients were considered at risk for hereditary status in the following situations: age at diagnosis <60 years with no previous history of bladder cancer; previous history of HNPCC-related cancer regardless of age; one first-degree relative with HNPCC-related cancer diagnosed before 50 years of age or two first-degree relatives diagnosed regardless of age.

RESULTS

  • • 
    Overall, 239 patients (21.3%) were considered to be at risk of hereditary UUT-UC.
  • • 
    Compared with sporadic cases, hereditary cases are more likely to be female (P= 0.047) with less exposure to tobacco (P= 0.012) and occupational carcinogens (P= 0.037). A greater proportion of tumours were located in the renal pelvis (54.5% vs 48.4%; P= 0.026) and were lower grade (40% vs 30.1%; P= 0.015) in the hereditary cohort.
  • • 
    The overall, cancer-specific and recurrence-free survival rates were similar in both cohorts.
  • • 
    We propose a patient-specific risk identification tool.

CONCLUSIONS

  • • 
    A significant proportion (21.3%) of patients with newly diagnosed UUT-UC may have underlying HNPCC as a cause.
  • • 
    Recognition of such potential and application of a patient-specific checklist upon diagnosis will allow identification and appropriate clinical and genetic management for patient and family.

Abbreviations
HNPCC

hereditary non-polyposis colorectal cancer

UUT-UC

upper urinary tract urothelial carcinoma

EAU

European Association of Urology

OS

overall survival

CSS

cancer-specific survival

RFS

recurrence-free survival

MFS

metastasis-free survival

MSI

microsatellite instability

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

Hereditary non-polyposis colorectal cancer (HNPCC), also known as Lynch syndrome, is an autosomal dominant multi-organ cancer syndrome caused by germline mutations of mismatch repair genes. This is the most common monogenetic cause of colorectal cancer, accounting for up to 7% of such tumours [1]. Cancers other than colorectal cancer (e.g. endometrial or ovarian cancer, cancer of the small bowel, stomach, pancreas, biliary tract, skin, brain or genitourinary tract) may be associated with HNPCC [2]. Upper urinary tract urothelial carcinomas (UUT-UCs) rank third (5%) after colon (63%) and endometrial cancer (9%) within the group of HNPCC-related tumours [3]. Patients with HNPCC disorder have a 6% lifetime risk of developing UUT-UC, which is 22 times higher than the risk seen in the general population [4]. Many urologists are not aware of this association, and it is presumed that some hereditary cancers are misclassified as sporadic, which results in underestimated incidence [4]. Consequently, family members of patients with UUT-UC secondary to HNPCC are denied appropriate surveillance and early detection.

The diagnosis of HNPCC is usually suspected on the basis of clinical criteria and confirmed by genetic testing. The Amsterdam Criteria II represent the most widely accepted clinical guidelines for inclusion as part of the HNPCC syndrome [5]. However, the application of such stringent criteria for diagnosis has a sensitivity of only 72% [6]. We have already demonstrated that the use of more inclusive criteria to pinpoint patients who are likely to have a hereditary tumour could be useful to identify HNPCC syndrome in UUT-UCs [4,7]. In addition, the Bethesda guidelines were published to help in the decision process and were aimed at determining which individuals should undergo genetic testing (Table 1) [8]. Based on the most recent European Association of Urology (EAU) guidelines on UUT-UC, a hereditary tumour has to be suspected if the patient is less than 60 years of age or has a personal or family history of an HNPCC-associated cancer [9]. The aim of this study was to analyse the demographic and epidemiological factors that are associated with a suspicion of hereditary UUT-UC in a large cohort from a national collaborative group.

Table 1. Revised Bethesda criteria (2003)
1. Diagnosed with colorectal cancer before the age of 50 years
2. Synchronous or metachronous colorectal or other HNPCC-related tumours (which include stomach, ureter, renal pelvis, biliary tract, glioblastoma, sebaceous gland adenoma, keratocanthoma and carcinoma of the small bowel), regardless of age
3. Colorectal cancer with a high-microsatellite-instability morphology that was diagnosed before the age of 60 years
4. Colorectal cancer with one or more first-degree relatives with colorectal cancer or other HNPCC-related tumours. One of the cancers must have been diagnosed before the age of 50 years (this includes adenoma, which must have been diagnosed before the age of 40 years)
5. Colorectal cancer with two or more relatives with colorectal cancer or other HNPCC-related tumours, regardless of age

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

STUDY DESIGN AND POPULATION

A multi-institutional, national, collaborative retrospective study was performed. We searched for suspicious cases of hereditary tumour among the 1122 patients from the French national database who were diagnosed and treated for UUT-UC between 1995 and 2010. For each patient, the database included the following parameters: gender, age at diagnosis, smoking status with the duration of intoxication, professional exposure to carcinogens (aromatic amines, polycyclic hydrocarbons, chlorinated solvents, arsenic), personal and familial history of tumour from the HNPCC-associated cancer spectrum (endometrial or ovarian cancer, cancer of the small bowel, stomach, pancreas, biliary tract, skin, brain or genitourinary tract), circumstances of discovery (gross haematuria, pain, alteration in general condition, acute renal failure, overactive bladder symptoms, follow-up after bladder cancer, incidental finding), tumour characteristics (location, TNM stage, 1973 WHO grade and lymph node status), follow-up and oncological outcomes.

CLINICAL CRITERIA FOR SUSPICION OF A HEREDITARY UUT-UC

Patients were considered at risk for hereditary UUT-UC if they presented at least one of the following clinical criteria, which we published previously [4,7]: age at diagnosis of <60 years with no previous history of bladder cancer; personal history of any other HNPCC-related cancer regardless of age; one first-degree relative with HNPCC-related cancer diagnosed before 50 years old or two first-degree relatives with HNPCC-related cancer regardless of age [8].

PATHOLOGICAL EVALUATION

All surgical specimens were examined by dedicated genitourinary pathologists. Tumours were staged according to the International Union Against Cancer TNM classification. Tumour grading was assessed according to the WHO guidelines of 1973. In cases of multifocality (ureteric and renal pelvic tumours), the lesion with the worst tumour characteristics was considered the index case.

FOLLOW-UP AND ONCOLOGICAL OUTCOMES

Patients were followed up at 3 and 6 months, and then every 6 months for 3 years after surgery and annually thereafter. Follow-up consisted of a physical examination, blood laboratory tests, urinary cytology, cystoscopic evaluation of the urinary bladder and a thoraco-abdomino-pelvic CT scan. Disease recurrence was defined as any documented tumour in the ipsilateral retroperitoneum, regional lymph nodes, bladder and contralateral UUT. A tumour located in the systemic organs was defined as metastasis. The cause of death was corroborated by case notes and death certificate review.

STATISTICAL REVIEW

The clinicopathological features in the two groups were compared using χ2 tests for categorical variables and Mann–Whitney tests for continuous variables. Survival was estimated using the Kaplan–Meier method and compared among groups with the log-rank test. Oncological outcomes focused on overall survival (OS), cancer-specific survival (CSS), recurrence-free survival (RFS) and metastasis-free survival (MFS). Patients were censured at last follow-up or death if the event of interest had not been attained. All tests were two-sided, with P < 0.05 considered significant. Statistical analysis was performed using XLSTAT Version 2011.3.01 (Addinsoft, Paris, France).

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

POPULATION

According to the criteria selected, 239 patients (21.3%) were considered at risk of hereditary UUT-UC, while 883 patients (78.7%) were assumed to have a sporadic tumour. Among patients at risk, 79 (33%) had a personal history of another HNPCC-associated cancer, 14 patients (5.9%) had a first-degree relative diagnosed with HNPCC-associated cancer before 50 years of age and six patients (2.5%) had two first-degree relatives with HNPCC-associated cancer regardless of age; 169 patients (70.7%) were under 60 years old with no previous history of bladder cancer. The overall male to female ratio was 2.28:1 but this difference was significantly less in the hereditary group (1.8:1, P= 0.047) due to an increased proportion of female patients. The overall median age at diagnosis was 70.4 years (range 61.9–76.9). Logically, due to our selection criteria, the patients selected as at risk for hereditary UUT-UC were younger (56.1 ± 11.9 years vs 72.5 ± 8.6 years) and hereditary tumour was more often diagnosed after visible haematuria (59.2% vs 49.5%, P= 0.008) or incidentally (19.1% vs 10.1%, P < 0.001). Environmental risk factors of UUT-UC were represented by former or active tobacco intoxication for 64.2% of patients and occupational exposure to chemical carcinogens (aromatic amines, polycyclic hydrocarbons, chlorinated solvents, arsenic) in 22.7%. Subgroup analysis showed environmental risk factors were significantly less important in the hereditary group as exposure to tobacco and occupational carcinogens was less prevalent (56.7% vs 66.4%; P= 0.012, and 13.9% vs 25.9%; P= 0.037, respectively).

TUMOUR CHARACTERISTICS

Overall, 371 patients (37.5%) had ureteral tumours, and 492 patients (49.8%) had renal pelvic tumours. The remaining 125 cases (12.7%) were multifocal tumours. Compared with sporadic cases, the location of the tumours in patients at risk of hereditary UUT-UC was significantly different. Ureteric, renal pelvic and multifocal tumours occurred in 30%, 54.5% and 15.5% of hereditary tumours compared with 39.7%, 48.4% and 11.9% in the sporadic group (P= 0.026). There was no difference in the TNM stage of tumours in the two groups. The lymph node status and the metastatic status at the time of diagnosis were similar in the two groups with 25.3% vs 26.7% of patients classified as N+ (P= 0.885) and 13% vs 11.8% of patients classified as M+ (P= 0.615), respectively. However, the presence of low grade disease was more frequent in the group at risk of hereditary UUT-UC (40%) than in the sporadic group (30.1%) (P= 0.015). Table 2 displays all relevant demographic and pathological data.

Table 2. Demographic and pathological characteristics of patients with UUT-UC
VariablesAll patients,n= 1122Patients at risk for hereditary HNPCC, n= 239Patients with sporadic UUT-UCC, n= 883 P
  • *

    Significant P values.

Age, years, median (range)70.4 (26–100)56.1 (26–87)72.5 (44–100)<0.001*
Sex (%)   0.047*
 Male774 (69.5)153 (64.3)621 (71) 
 Female339 (30.5)85 (35.7)254 (29) 
Smoking (%)   0.012*
 Never324 (35.8%)87 (43.3)237 (33.6) 
 Current or former582 (64.2%)114 (56.7)468 (66.4) 
Professional exposure to carcinogens (%)   0.037*
 No214 (77.3)62 (86.1)152 (74.1) 
 Yes63 (22.7)10 (13.9)53 (25.9) 
Circumstances of discovery (%)    
 Haematuria556 (51.6)138 (59.2)418 (49.5)0.008*
 Pain383 (35.5)85 (36.6)298 (35.1)0.673
 Alteration in general condition110 (10.2)27 (11.5)83 (9.8)0.443
 Acute renal failure65 (6.1)10 (4.4)55 (6.6)0.211
 Overactive bladder symptoms89 (8.2)16 (6.9)73 (8.6)0.398
 Incidental finding129 (12)44 (19.1)85 (10.1)<0.001*
pT (%)   0.554
 CIS26 (2.58)3 (1.33)23 (2.93) 
 pTa283 (28)62 (26.7)221 (28.1) 
 pT1204 (20.2)51 (22.8)153 (19.5) 
 pT2107 (10.6)20 (8.93)87 (11.1) 
 pT3311 (30.8)68 (30.4)243 (31) 
 pT478 (7.73)20 (9.93)58 (7.39) 
pN (%)   0.885
 pN0282 (73.6)71 (74.7)211 (73.3) 
 pN+101 (26.4)24 (25.3)79 (26.7) 
pM (%)   0.696
 pM0987 (88)208 (87)779 (88.2) 
 pM+135 (12)31 (13)104 (11.8) 
Grade (%)   0.015*
 Low grade243 (32.4)68 (40)175 (30.1) 
 High grade508 (67.6)102 (60)406 (69.9) 
Location (%)   0.026*
 Ureteral tumours371 (37.5)66 (30)305 (39.7) 
 Renal pelvic tumours492 (49.8)120 (54.5)372 (48.4) 
 Multifocal tumours125 (12.7)34 (15.5)91 (11.9) 

SURVIVAL

The mean follow-up was 33.1 months (median 24.5; range 9.3–48.7). Overall, 263 patients (23.4%) died during follow-up including 203 (18.1%) from UUT-UC. The 5-year OS rate was 64% (95% CI 59.8–68.1), and the 5-year CSS rate was 71.6% (95% CI 67.7–75.4) (Fig. 1). There was no significant difference between the group at risk for hereditary UUT-UC and the sporadic cases (P= 0.246 for OS and P= 0.401 for CSS). Overall, 430 patients (38.3%) experienced recurrence during the follow-up. The 5-year RFS rate was 45.4% (95% CI 41.2–49.6), and the two groups were comparable (P= 0.067) (Fig. 1). Metastases occurred in 236 patients (21%). The median time for metastases development was 8.5 months. The 5-year MFS probability was 58.3% (95% CI 53.9–63.3) with no difference between the two groups (P= 0.838).

image

Figure 1. A, Comparing overall survival curves in patients with sporadic (green) or suspected hereditary (red) UUT-UC shows no difference (P= 0.246). B, Comparing cancer-specific survival curves in patients with sporadic (green) or suspected hereditary (red) UUT-UC shows no difference (P= 0.401). C, Comparing recurrence-free survival curves in patients with sporadic (green) or suspected hereditary (red) UUT-UC shows no difference (P= 0.067).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

Although UUT-UCs are rare, with an annual estimated incidence in western countries of approximately one or two new cases per 100 000 inhabitants [9], epidemiological data demonstrate that in patients with HNPCC syndrome the cumulative risk of UUT-UC ranges up to 6% [3]. Watson et al. [3] reported a 22-fold increased risk compared with the general population. This may even represent an underestimation as the authors did not consider UUT-UCs of the renal pelvis together with those of the ureter. To our knowledge, there are few publications investigating the risk of having an HNPCC tumour among patients with UUT-UC [4,7].

It has already been postulated that some hereditary cancers are misclassified as sporadic and that their incidence is underestimated in the general population [4,10]. Patients who are suspected of having a hereditary syndrome present different demographic and epidemiological characteristics from patients with sporadic tumours. In the current study, patients at risk of a hereditary tumour had less personal exposure to carcinogens (P= 0.012) and, although UUT-UC remains more common in men, the male to female ratio decreases significantly (P= 0.047). In addition, suspected hereditary tumours were more likely to be located in the renal pelvis. Our results are in line with the study by Crockett et al. [10], who described the characteristics of UUT-UC among 39 patients with a diagnosis of HNPCC: median age at diagnosis was 62 years vs 70 in the general population, with a mean of 15.85 years after prior HNPCC-associated cancer. Only half of the patients had a significant history of smoking, and the female to male ratio was 1.05:1. Tumour grade and outcomes were similar to those of the general population. Contrary to our results, 51% of patients with HNPCC had ureteric UC. A predominance of ureteral lesions in patients with high microsatellite instability (MSI) was also found in another study [7].

As already published for colorectal cancer, application of the stringent Amsterdam clinical criteria for the diagnosis of HNPCC reveals that 6% of colorectal cancers are hereditary [1]. HNPCC syndrome is a consequence of genetic mutation in one of the DNA mismatch repair genes (MLH1, MSH2, PMS2 and MSH6) [1]. Subsequently, the ability to repair mutations at the nucleotide level is lost and instability of the genome, termed MSI, is used as a marker of predisposition to the increased mutation rate that facilitates the development of cancer [11]. MSI can help detect germline mutations, allowing for the detection of potentially hereditary cancer [7,8,12]. In a former study, we have already shown that MSI testing was found to be positive in 40% of 164 patients diagnosed with UUT-UC, with a high level of instability for 27 patients (16.5%). Among them, 11% had a mutation of MSH2, which in turn was significantly related to a personal history of HNPCC-related cancer and the occurrence of UUT-UC before the age of 60 [7,13,14].

Thus, the 2011 EAU guidelines emphasize that patients with a newly diagnosed UUT-UC should be suspected of a familial or hereditary tumour linked to HNPCC and a thorough tailored medical interview is necessary. Patients presenting at <60 years of age or with a personal and/or familial history of HNPCC-related cancer are suspected to have hereditary aetiology [9]. Recently, Archer et al. [15] proposed a rational strategy to screen young patients with UUT-UC but no known diagnosis of HNPCC. The approach involves screening tumour tissue to identify MSI if patients are under the age of 60 years with no previous history of bladder cancer. The Bethesda criteria were published to decide which individuals with suspicion of hereditary cancer, but not fulfilling the Amsterdam criteria, should undergo genetic testing [8]. Several studies showed that the Bethesda criteria were more sensitive than the Amsterdam (94% vs 72%) [6,16].

In our study, strikingly 21.3% of 1122 patients have a suspected hereditary UUT-UC based on these reference criteria. This has significant implications for both the patient and, arguably more pertinently, for family members. Based on our study findings of equivalent OS, CSS and RFS in patients with sporadic and suspected hereditary tumours, the focus of attention may largely rest upon family members. Upon diagnosis of UUT-UC in an at-risk individual, family members would undergo thorough assessment including medical history, physical examination and screening for tumours belonging to the HNPCC spectrum (e.g. UUT-UC, colorectal carcinoma, uterine cancer) [17]. Obviously, potential for early diagnosis is always higher in a population that is pre-informed of a higher risk of disease compared with the general population. However, for personal and holistic reasons, it still remains important to inform individual patients with newly diagnosed UUT-UC if they are considered to be at risk of an underlying hereditary aetiology. In Table 3, we propose a patient-specific checklist to be utilized by the attending urologist upon diagnosis of UUT-UC in all patients. Such a tool allows efficient identification of potential hereditary UUT-UC and referral of the patient and/or family members for genetic testing and counselling and will stop misclassification of UUT-UC as sporadic.

Table 3. Proposed risk assessment tool to identify patients newly presenting with UUT-UC who may have an underlying hereditary aetiology
Age at diagnosis<60 years>60 years
SexFemaleMale
Personal history HNPCC-associated cancer
 Tick if yes
Colorectal cancer
Stomach cancer
Biliary cancer
Contralateral UUT-UC
Endometrial cancer
Ovarian cancer
Small bowel cancer
Glioblastoma
Family history HNPCC-associated cancer
 Tick if yesTick if 1st degreeTick if <50
Colorectal cancer
Stomach cancer
Biliary cancer
Contralateral UUT-UC
Endometrial cancer
Ovarian cancer
Small bowel cancer
Glioblastoma
Suspicion of hereditary UUT-UC (if tick any of below then suspect)
Age <60
Personal history of HNPCC-associated cancer
First-degree relative <50 with HNPCC-associated cancer
Two first-degree relatives with HNPCC-associated cancer

We would like to address a few limitations of our epidemiological study. We acknowledge that the lack of genetic testing to confirm definitively hereditary UUT-UC is the key limitation. Due to the scarcity of UUT-UC cases, the 1122 patients in our database have been accrued over 15 years from 17 French institutions. Sadly, it would be a logistical impossibility to perform mismatch repair gene testing on the 239 patients we considered at risk for hereditary UUT-UC. Second, it could be argued that our choice of age <60 years as a risk factor for hereditary UUT-UC is arbitrary. However, we chose this cut-off as it is the age quoted by the recent European guideline committee on UUT-UC [9], is in line with the median age stated in other series of hereditary UUT-UCs [10] and was identified as an age cut-off previously by our group [4,9]. Lastly, the retrospective nature of our database analysis, although unavoidable, brings with it all the inherent flaws of such analysis. However, we feel we have presented the largest cohort of UUT-UC patients investigated with reference clinical criteria to determine the suspicion of underlying hereditary UUT-UC secondary to possible HNPCC syndrome.

In conclusion, using reference clinical criteria for the diagnosis of HNPCC syndrome, we have shown from analysis of a multi-institutional database that a significant proportion (21.3%) of patients with newly diagnosed UUT-UC may have underlying HNPCC as a cause and are misclassified as sporadic UUT-UC. Recognition of such potential and application of a patient-specific risk identification tool upon diagnosis will allow appropriate clinical and genetic management for both the patient and family members.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES