To report our experience with ureteroscopic laser ablation of upper tract urothelial carcinoma (UTUC) in patients with Lynch Syndrome (LS), as defined by a documented germline mutation in the MSH-2 gene.
To increase awareness among urologists about UTUC in this unique patient population and refer to genetic counselling when appropriate.
Patients and Methods
Demographic, clinical and pathological data on 13 consecutive patients with UTUC and documented MSH-2 mutation comprising 15 involved renal units were retrospectively collected.
Ureteroscopic evaluations involved biopsy and laser treatment with combination holmium/neodymium yttrium aluminum garnet (YAG) lasers.
Tumours were graded from 1 to 3 according to the 1973 World Health Organisation classification by a single pathologist evaluating cell block preparations.
The mean patient age at initial presentation was 56.5 years, with six of 13 patients having metachronous bilateral UT disease.
The mean follow-up was 59 months with a mean number of surveillances of 12.
Of 15 affected renal units, 10/15 (67%) of initial tumours involved the ureter with mean lesion size of 17.5 mm, while five of 15 (33%) involved the intrarenal collecting system with mean lesion size of 25 mm.
Ureteroscopy cleared 13/15 (87%) lesions and four of those 13 (31%) needed staged procedures. Renal preservation rate was 14/15 (93%) with one nephroureterectomy and one segmental ureterectomy performed.
One patient developed metastatic UTUC after 40 months surveillance. No patient presented with bladder tumours but seven of the 13 (54%) developed them within 10 months of the initial ureteroscopy.
Patients with LS who develop UTUC present at younger ages and appear to be more likely to have bilateral UT disease over their lifetimes vs sporadic UTUC patients.
Ureteroscopic laser ablation offers a good renal preservation rate with reasonable cancer control in patients willing to undergo endoscopic surveillance.
Endoscopic management of upper tract urothelial carcinoma (UTUC) has emerged as an important alternative to nephroureterectomy (NU) in select patients. Originally reserved for patients with imperative indications, several studies have shown ureteroscopic treatment of low-to-intermediate grade superficial UTUC lesions to be a safe alternative to NU in carefully selected patients willing to undergo periodic surveillance, including those with normal contralateral kidneys [1, 2]. Sporadic cases of UTUC usually affect individuals within the seventh to eighth decades of life and are predominantly unilateral. Lynch Syndrome (LS), also known as hereditary nonpolyposis colorectal cancer (HNPCC), is an autosomal dominant cancer syndrome characterised by mutations in DNA mismatch repair (MMR) genes. Those MMR genes most often mutated in carriers of a LS trait are MSH-2, MLH-1, MSH-6 and PMS-6 . LS carriers are prone to various malignancies including colorectal, endometrial and UTUC. Recently, a case-control study showed that patients with LS are diagnosed with UTUC at younger ages compared with a control series of sporadic UTUC cases . Furthermore, a relative risk of 75 for developing UTUC of the renal pelvis or ureter was found in patients with known MSH-2 mutations compared with the general population . Small numbers and an uncertain natural history for the incidence of bilateral disease make absolute treatment recommendations for UTUC in patients with LS difficult to offer. The present series of 13 consecutive patients with documented MSH-2 mutations and UTUC is presented as an example of predominantly endoscopic treatment for this unique condition. An additional goal of the present study is to increase awareness among urologists of UTUC in patients with HNPCC, as a recent retrospective study of the French collaborative national database on UTUC suggested that ≈21% of patients newly diagnosed with UTUC, originally classified as sporadic cases may actually be hereditary in nature, thus having not only implications for patients but also their family members .
Patients and Methods
After obtaining Institutional Review Board approval, a retrospective chart review of all patients who presented to our institution from 1995 to 2010 and had at least one ureteroscopic evaluation for UTUC was undertaken to identify those with an established diagnosis of HNPCC trait carrier as determined by genetic mutation analysis. This review comprised a 16-year experience involving 466 patients in our referral practice. Patients were usually identified initially because they met Amsterdam Criteria II; they were either sent to us with a documented germline MSH-2 mutation or they were sent for genetic counselling and mutational analysis by our group. Germline MMR gene testing consisted of full sequence determination of all introns and immediately adjacent intronic regions of the MLH-1, MSH-2 and MSH-6 genes and multiplexed quantitative PCR of MLH-1 and MSH-2 to assess for deletion/duplication mutations.
Amsterdam Criteria II
Requires at least three relatives with an HNPCC associated cancer: colorectal cancer; cancer of the endometrium, small intestine, ureter or renal pelvis; all of the following criteria should be present:
One should be a first-degree relative of the other two;
At least two successive generations should be affected;
At least one should be diagnosed before age 50 years;
Familial adenomatous polyposis (FAP) should be ruled out in any present cases of colorectal cancer;
Tumours should be verified by pathological examination.
Patient charts were reviewed to determine age at diagnosis with UTUC, frequency of bilateral UT disease, tumour location, tumour grade and stage when applicable, number of surveillances, months of follow-up, disease progression, recurrence and presence of additional primary cancers. Attention was directed to the type of healthcare provider who initiated a genetics consultation and the time from initial diagnosis of an HNPCC-related cancer to the time a genetics consultation was obtained.
Some patients had previous surgery at other institutions including nephroureterectomies and segmental ureterectomies. These four renal units were treated prior to patient presentation to us and thus are excluded from outcomes analysis for ureteroscopic treatment at our institution. Ureteroscopic evaluations were performed on all patients and all UT tumours were biopsied and treated with a combination of neodymium:yttrium aluminum garnet (YAG) laser for coagulation and holmium:YAG laser for ablation/resection as previously described . Biopsies and site-specific urine aspirates were prepared and processed by our cytopathology laboratory and evaluated using a previously described technique . Cell block preparations were made whenever sufficient visible tissue was available. Tumours were graded from 1 to 3 according to the 1973 WHO classification by a single pathologist evaluating cell block preparations. To allow comparison of pathological grading to contemporary series, we retrospectively applied the 2004 WHO grading system for UCs to our initial ureteroscopic biopsy samples for each of the 15 renal units treated with ureteroscopy.
Primary tumours that were large or difficult to treat completely due to bleeding or poor visibility on initial ureteroscopy were completed with staged procedures. Patients were followed on a surveillance protocol involving cystoscopy, retrograde pyelogram and ureteroscopy with laser ablation every 3 months until free of tumour. Once the patient was free of any viable tumour, ureteroscopic evaluation was performed every 6 months for 5 years, and yearly thereafter. If the contralateral renal unit is uninvolved, it is evaluated by retrograde pyelogram every 6 months or yearly after the first 5 years. CT or MRI of the abdomen and pelvis with contrast were ordered if there was any change in baseline retrograde pyelogram suggesting incomplete treatment or local advancement of disease. To qualify for continued endoscopic treatment and surveillance, pathology had to be low-to-intermediate grade and all tumours in the intraluminal collecting system had to be accessible by semi-rigid or flexible ureteroscopy. In the vast majority of cases, patients with high-grade lesions or inaccessible tumour and a normal contralateral kidney were treated with laparoscopic NU or segmental ureterectomy as indicated.
In all, 13 patients with UTUC and positive genetic mutation analysis were identified. All 13 patients had a mutation in the MSH-2 gene confirmed after meeting with a genetic counsellor or medical oncologist specialising in medical genetics. Patient demographics are given in Table 1. Taking into account patient history before consultation with our institution, six of the 13 patients developed bilateral disease over time and all were metachronous with notable variation in the interval to bilateral disease development ranging from 1 month to 120 months. Four of these patients had extirpative surgery before presenting to our department with three having radical NU and one having a segmental ureterectomy. Thus, we had three patients come to us with solitary kidneys and the patient with a segmental ureterectomy has never had recurrence in that renal unit. The two other patients with bilateral disease were discovered after presenting to our department, therefore this ureteroscopic experience involves 13 patients with 15 affected renal units (Fig. 1).
Table 1. Demographics of 13 patients with HNPCC with UTUC and documented MSH-2 mutation
Mean (range) age, years
Diagnosis of first UTUC
Diagnosis of first HNPCC cancer
Initial meeting with genetics counsellor
Significant medical history, n/N:
Pelvic radiation therapy
Positive family history of colorectal cancer
Positive smoking history
Nephron sparing indications, n/N:
Chronic renal insufficiency
Bilateral disease (no history of NU)
History of bilateral disease. n/N:
Mean (range) time to bilateral disease development, months
Initial treatment in those with bilateral disease, n/N:
History of radical NU
History of segmental ureterectomy
Ureteroscopy with laser ablation
Successful ureteroscopic clearance of the initial UTUC lesion was achieved in 13 of the 15 (87%) renal units (Table 2). Of those cleared successfully, nine of 13 were able to be completely treated in one session, while four of 13 required staged procedures for complete treatment. Three of the four cases requiring staged procedures were intrarenal neoplasms measuring ≥2.5 cm. Two patients were unable to be completely cleared endoscopically. One of these patients required a formal distal ureterectomy for a 4-cm long, high-grade ureteric tumour. Final pathology of the resected ureteric segment after attempted ureteroscopic laser ablation showed focal high-grade dysplasia with no invasive carcinoma detected. The second patient unable to be cleared endoscopically was the only patient to go on to develop metastatic disease from UTUC in the series, had a solitary kidney, and declined radical NU. She presented with a 3-cm low-grade renal lesion, which ultimately progressed to a high-grade lesion 29 months after initiating ureteroscopic treatment. At 40 months follow-up she had a high-grade bladder lesion and retroperitoneal lymphadenopathy. She is actively getting palliative chemotherapy.
Table 2. Ureteroscopic treatment and surveillance outcomes in 13 HNPCC patients with UTUC (15 renal units)
Number of surveillances
Initial lesion cleared with ureteroscopy:
Cleared in one procedure
Cleared with staged procedures
Initial lesion not cleared with ureteroscopy:
Treated with segmental ureterectomy
Treated with palliative ureteroscopy
Renal preservation rate
Radical NU (non-functioning kidney)
Local UT recurrence rate (for units still at risk)
Grade progression of UT lesion
Bladder tumour incidence:
Prior to UTUC treatment
After UTUC treatment
Metastatic disease development:
From other primary cancer
Ureteric stricture development
Tumour characteristics are detailed in Table 3. For initial tumour location, lesions involved the ureter in 10/15 (67%) cases and the mean (range) size was 17.5 (5–40) mm. Initial lesions involved the intrarenal collecting system in five of 15 (33%) and the mean (range) size was 25 (15–30) mm. Initial tumours were low-to-intermediate grade in 13/15 (87%), while two of 15(13%) were high grade using the 1973 WHO classification. One patient with high-grade disease had segmental ureterectomy performed as described above. The second patient presenting with high-grade disease had a 10-mm mid-ureteric lesion with a pre-existing ureteric stent. This lesion was easily cleared ureteroscopically and after discussion with the patient, he remained on ureteroscopic surveillance. This patient never went on to have a recurrence in the UTs or bladder and is still disease free at 76 months follow-up and his contralateral kidney is completely normal. Of note, when the 2004 WHO classification was retrospectively applied to initial biopsy specimens, 11/15 (73%) were found to be low grade while four of 15 (27%) were high grade. Thus, two grade 2 lesions by 1973 criteria were retrospectively judged to be high grade by 2004 criteria.
Table 3. Tumour characteristics of UTUC in 13 HNPCC patients (15 renal units) with documented MSH-2 mutation treated with ureteroscopy
Initial tumour location, n/N (%)
Mean (range) initial tumour size, mm
Initial tumour grade (1973 system), n/N (%):
Initial tumour grade (2004 system), n/N (%)
In the present series, our 13 patients with 15 affected renal units undergoing ureteroscopic surveillance according to the above protocol had a mean (range) number of surveillances of 12 (5–36) per patient with a mean follow-up of 59 (10–188) months (Table 2). As mentioned above, during ureteroscopic surveillance, only one of 13 patients had disease progression from a low-grade lesion in the renal pelvis to a high-grade lesion. In the 13 patients whose initial lesion was successfully cleared with ureteroscopy, three developed metachronous ipsilateral UT recurrences, which were generally small and easily treated with ureteroscopy. Time to recurrence was variable among the three patients at 3, 12 and 177 months.
During endoscopic surveillance, we noted that seven of the 13 of patients developed a metachronous bladder tumour and six of these seven manifested within 10 months of initial surveillance. High-grade invasive bladder tumours were noted in two of the seven patients. No patient had urothelial cell carcinoma of the bladder preceding treatment of an UT tumour. Most patients had additional primary cancers (11of the 13), most of which included colon or endometrial cancer.
Ureteric strictures rendering patients as stent dependent were seen in four of the 15 (27%) renal units and three of these four had prior history of pelvic radiation, while two of these four had primary ureteric tumours. One patient developed a stricture after five ureteroscopic surveillances who did not have a primary ureteric lesion or history of pelvic radiation. Of note, despite not having any visualised tumour recurrences, this patient had persistently positive urine cytology, received intraluminal UT mitomycin C (MMC) as an outpatient and thus had multiple ureteric catheterisations in addition to ureteroscopic surveillances. This patient went on to develop a non-functioning kidney and had a laparoscopic NU performed. Taking this case into account, our nephron-sparing rate was 14/15 (93%). Final pathology on this NU specimen showed only urothelial dysplasia. Overall, two of the 13 patients in the series died. One patient died of metastatic colorectal cancer; the other died from metastasis of duodenal primary adenocarcinoma. While using our protocol, only one patient developed metastatic UTUC as described above.
Upon initial presentation, six of the 13 patients presented with UTUC while seven of the 13 manifested with a different HNPCC-associated cancer initially (Table 4). Of the 19 renal units affected with UTUC throughout the experience, including those 4 units undergoing extirpative surgery elsewhere, 12/19 (63%) patients presented with gross haematuria. Four of 19 (21%) were found incidentally with imaging. Three of these were in asymptomatic patients getting surveillance CT scans for other previously diagnosed cancers. Of note, six of the 13 patients presented to us with UTUC and a previously determined genetic diagnosis. Four of these had colon or endometrial cancer while one had metachronous bilateral UTUC only and one had only a sebaceous adenoma of the skin. Our group sent the remaining seven patients for genetic testing. Of these, three were first-degree relatives of one another, three others had metachronous bilateral UTUC and one had personal history of colorectal cancer before UTUC. Family history for colorectal cancer was seen in the vast majority of the present patients (12/13). The time from initial HNPCC cancer diagnosis to genetics consult was very variable and ranged from 1 to 276 months.
Table 4. Initial presentations of 13 HNPCC patients with UTUC and documented MSH-2 mutation
Initial HNPCC cancer diagnosis
Colorectal carcinoma (CRC)
Endometrial carcinoma (EMC)
Simultaneous UTUC and CRC
Simultaneous CRC and EMC
Sebaceous adenoma (skin)
Initial presenting signs/symptoms of UTUC (19 renal units):
Found on screening studies
Specialist sending patient for genetic counselling:
Mean (median, range) time from initial HNPCC cancer diagnosis to genetics consult, months
73 (12, 1–276)
It has been estimated that malignant UTUCs account for 5–10% of all renal neoplasms and ≈5% of all UCs. Despite being a relatively rare entity, disease-specific survival in patients with UTUC is clearly influenced by tumour stage . Historically, preoperative staging in these patients has been challenging but many groups have shown that tumour grade predicts, with good accuracy, tumour stage [10-12]. With the development of improved flexible ureteroscopes and laser technology have come effective nephron-sparing ablative treatments for well-selected patients with UTUC, including those with normal contralateral kidneys [1, 2].
Although relatively rare in the general population, UTUC has been found to be part of the tumour spectrum in patients with LS , particularly in those patients with MSH-2 mutations . The lifetime risk of UTUC in patients with LS is thought to approach 6%  and heterogeneity seems to exist among different families with HNPCC . Multiple retrospective studies have suggested LS patients with UTUC present at ages 7–9 years younger than sporadic cases of UTUC [13, 15, 16]. Only recently has there been a systematic comparison of UTUC patients with LS with UTUC versus sporadic UTUC cases . This case-control study showed that patients with LS presented with UTUC 8 years younger than those with sporadic UTUC and that patients with LS were more likely to have ureteric involvement compared with renal pelvis involvement. By comparison, the present cohort of LS-trait carriers was also relatively young at the time of presentation with a mean age of 56.6 years (Table 5). We also noted a more frequent initial involvement of the ureter (10/15, 67%) compared with the intra-renal collecting system (five of 15, 33%).
There were similar numbers of pathological high-grade cases in the LS and sporadic patients in the Crockett et al.  case-control study (88% and 74%, respectively). Of note, these cohorts were all treated with NU or segmental ureterectomy and comprised a 42-year experience from 1964 to 2006 . By contrast, most of the present patients had low- or intermediate-grade disease on initial ureteroscopic biopsy (13/15, 87%). This discrepancy in pathological grading among series may have been due to our use of the 1973 WHO classification system for grading UCs as opposed to the 2004 WHO criteria used by Crockett et al. . We feel the former grading system allows for greater flexibility in diagnosing tumour heterogeneity. We also made use of this system to maintain consistency throughout the present study, as our first patient in this series was diagnosed in 1995, before the availability of the 2004 classification system. However, the 1973 classification may lead to more intermediate-grade cases and fewer high-grade cases due to the heterogeneity of grade 2 lesions according to 1973 criteria [17, 18]. Nevertheless, when we retrospectively applied the 2004 classification system to our original ureteroscopic biopsy specimens, 11/15 (73%) were defined as low-grade lesions and only four (27%) were high grade. The sustained discrepancy in pathological grading among the present series and the Crockett et al.  study may be due to three other factors. First, both the present series and that of Crockett et al.  are relatively small containing 13 and 39 patients, respectively. Second, selection bias in our referral pattern exists as our group tends to see cases referred with absolute or relative indications for nephron-sparing endoscopic management in favour of extirpative options. Despite this, our initially treated lesions had a mean size of 17.5 mm in the ureter and 25 mm in the renal pelvis. Third, our grading was dependent on smaller biopsy samples compared with larger samples from extirpative procedures making undergrading due to sample error a possibility. Taken together, these factors make generalizations about the grade of UTUC lesions in patients with LS difficult.
No mention has been made previously in any study about the incidence of bilateral UT involvement in LS patients with UTUC and this is still not well defined. Considering patients' experience before treatment at our institution, there was bilateral UT involvement in six of the 13 patients in the present series with a mean (range) interval of 49 (1–120) months to contralateral disease development. This is in contrast to sporadic cases of UTUC, which have been reported to demonstrate metachronous bilateral involvement at a rate of 3.1%, as seen in a large series of 768 patients with a median (range) interval of 46 (2–232) months to contralateral disease development . Synchronous bilateral UTUC has been described in another large series of patients with UTUC at a rate of 2.7% . Selection bias might be responsible for our high rate of bilateral disease but having a genetic predisposition to UTUC may predispose patients with LS to a higher rate of bilateral involvement, especially in patients presenting younger. If this is indeed true, good conservative endoscopic management in favour of NU, when appropriate, may help avoid chronic renal insufficiency, similarly to those patients undergoing partial nephrectomy in favour of radical nephrectomy for cortical lesions .
Although no patients initially presented with bladder lesions, there was development of lower tract lesions in seven of the 13 patients over time, with most being found ≤10 months of initial ureteroscopic treatment. This rate of bladder tumour recurrence is slightly higher than reported ureteroscopic series for UTUC in which patients without prior bladder tumours went on to develop them after initial ureteroscopic treatment 26–45% of the time and developed them within 13–29 months of initial treatment [7, 11, 22, 23]. In comparison, Matsui et al.  reported a 42% rate of bladder tumour development in patients without prior lower tract lesions at a median follow-up of 39 months after open or laparoscopic NU. Although there were few patients in the present study, there was a relatively high rate of new bladder lesions in a relatively short duration after initial ureteroscopic treatment. This suggests the potential for downstream tumour seeding of the bladder after ureteroscopic treatment of UTUC in these patients, especially since UCs of the bladder are not considered part of the tumour spectrum in HNPCC. The common finding of new bladder tumour development in this patient cohort raises the need for future consideration of postoperative intravesical MMC administration. In patients with no previous bladder cancer history undergoing NU for UTUC, O'Brien et al.  reported a single postoperative dose of intravesical MMC decreased bladder tumour development within 12 months to 16% vs 27% in untreated controls. The effect of intravesical MMC to reduce bladder lesion development after ureteroscopic tumour ablation is unknown but might be beneficial with little adverse effects. Therefore, this area clearly warrants further study.
The overall survival rate for patients in this series was 11/13 with mean follow-up of 59 months. Two patients in the series died; one from metastatic colon cancer and the other from metastatic adenocarcinoma of the duodenum. One patient did go on to develop metastatic disease from UTUC. This patient had a solitary kidney. Including this patient, our kidney-sparing rate was 14/15 (93%). Two patients required extirpative surgery after ureteroscopic management. One laparoscopic NU was performed in the series for a non-functioning kidney after proximal ureteric stricture development. Final pathology showed urothelial cell dysplasia. Another patient who underwent segmental ureterectomy for high-grade UTUC showed only focal high-grade dysplasia with no invasive carcinoma on final pathology. These two cases show the potential treatment effectiveness of ureteroscopic laser coagulation and ablation.
In conclusion, this series represents the first report of ureteroscopic treatment of UTUC in patients with HNPCC with confirmed MSH-2 mutations. As a group, they are considerably younger than sporadic UTUC cases. From the present data, it appears that these patients have a relatively high rate of metachronous bilateral disease, making conservative therapy with ureteroscopy more attractive, as to avoid chronic renal insufficiency in these younger patients. A potential disadvantage to ureteroscopy in this population is the risk of ureteric stricture formation given greater propensity for tumour involvement of the ureter vs the intra-renal collecting system and the often, simultaneous history of other primary pelvic malignancies and radiation treatment. Additionally, new bladder tumour development is common. Urologists should be reminded of the importance of obtaining a detailed family history in patients with UTUC. Those individuals who fit Amsterdam Criteria II (see above) , or those diagnosed with UTUC at a relatively young age should be considered at high risk for a germline mutation and considered for referral to a genetic counsellor.
Conflict of Interest
Demetrius H. Bagley is a consultant for Cook, Bark and Olympus/ACMI.