Viva Practice for the FRCS(Urol) and Postgraduate Urology Examinations, 2nd ed.

Chapter 4 Urothelial Cancer

David Mak, Muhammad Jamal Khan, Herman S Fernando, Ciaran Lynch and Lyndon Gommersall

Haematuria

Q. How common is haematuria?

A. Haematuria is a common finding. In the paediatric population, this is as low as 1% rising to 5% in young adults and 10% in patients over the age of 50 years. The prevalence of non-visible haematuria has been reported to be approximately 2.5% in men and between 8% and 11% in women [1].

Q. A 65-year-old woman is referred with non-visible haematuria. A midstream specimen of urine (MSU) has been sent for microscopy, culture and sensitivity. Microscopy shows a normal white cell count (WCC), 18 red blood cells (RBCs) per high-powered field and no growth has been found on culture. She has moderate irritative voiding lower urinary tract symptoms (LUTS) and is otherwise well. No other investigations have been performed. How would you assess this woman?

A. This woman would be seen in a 2-week wait haematuria clinic and assessed with a focussed urological history, examination and investigations, detailing her smoking and occupational history. Environmental carcinogen exposure confers a less than 1% risk of developing bladder cancer but is an important component of assessment. Blood pressure should be measured as well as a full abdominal and pelvic examination should be conducted. Investigations should include an MSU for culture and sensitivity, urine for cytology, routine bloods, an ultrasound scan (USS) of her renal tract and a flexible cystoscopy. Some institutions advocate performing a computed tomography urogram (CTU) as the first-line radiological investigation. Urinary molecular marker tests (e.g. NMP 22 and UroVysion) have been developed to improve diagnostic sensitivity but none of these markers are currently accepted in routine practice or clinical guidelines.

If a bladder cancer is found then further staging investigations are required in selected cases and she would be counselled for transurethral resection of bladder tumour (TURBT). If these tests are negative with high-grade malignant cells on urine cytology then a CTU should be performed if not already done (contemporary radiological practice clearly favours the CTU over intravenous urography [IVU]). If this does not reveal an upper tract tumour then an examination under anaesthetic, rigid cystoscopy with biopsies, selective ureteric urine sampling, retrograde studies and ureteroscopy should be undertaken.

Q. What is the most likely presentation of a patient with bladder cancer?

A. The most common presentation of bladder cancer is with visible painless haematuria.

Almost all patients diagnosed with bladder cancer will have had either visible or non-visible haematuria. The degree of bleeding is not proportional to disease stage. Storage voiding symptoms are a worrying feature and can occur in approximately 20% of patients with either bladder cancer or carcinoma in situ. Urological bleeding associated with urinary infection cannot solely be attributed to inflammation as a proportion of patients will have necrotic infected elements within the bladder tumour and There fore must be investigated with equal intensity. Patients who receive imaging or a cystoscopy for another reason make up the remainder of patients.

Q. What are the causes of haematuria?

A. The causes of haematuria can be divided into urological or nephrological, benign or malignant, visible or non-visible, or based upon the anatomical location of the bleeding (i.e. renal, ureteric, bladder, prostatic or urethral). Table 4.1 demonstrates the causes of haematuria based on the anatomical location.

Table 4.1 Causes of haematuria based on anatomical location

Q. What is the definition of haematuria?

A. Haematuria can be divided into visible haematuria (VH) (i.e. macroscopic or gross haematuria) or non-visible haematuria (NVH) (microscopic haematuria). NVH can be further sub-classified into symptomatic non-visible haematuria (s-NVH) (i.e. with voiding LUTS) or asymptomatic non-visible haematuria (a-NVH) (i.e. without voiding LUTS).

The definition of NVH varies between guidelines (as do the recommended investigations that are suggested), and these definitions are summarised in Table 4.2. The quantification of NVH depends on two key techniques: the sediment count and the chamber count. The sediment count involves spinning urine down in a centrifuge with the supernatant removed. The pellet of cells is then re-suspended in saline and examined under the microscope. The chamber count detects the number of RBCs per millilitre of urine.

The American Urological Association (AUA) definition of haematuria utilises the sediment count and defines three or more RBCs per high-powered field as abnormal. Most nephrologists will utilise the threshold of 5 RBCs per microlitre. The Journal of the American Medical Association (JAMA) defined microscopic haematuria in 1990 as >2-3 RBCs per high-powered field [2]. Campbell’s Urology defines haematuria as 3 or more RBCs per high- powered microscopic field. Table 4.3 shows the relationship between the dipstick positive result and the RBC count per high-powered field. A trace result is generally considered negative and 1+ and above a positive result.

Table 4.2 Summary of varying definitions of non-visible haematuria

Table 4.3 Equivalent dipstick and microscopy results for non-visible haematuria

Khadra et al. has also reported on the percentage of patients who would have had a cancer diagnosis missed for 10, 5 and 3 RBCs per high-powered field. This equates to a missed cancer diagnosis in 20%, 14.8% and 10%, respectively [3].

Q. What do you know about dipstick testing for haematuria?

A. Dipstick testing for haematuria is commonplace. It is based on the oxidation of a chromogen (orthotolidine) by the peroxidase activity of haemoglobin. This results in a colourific change that is compared to a set of known standards. Electronic strip readers remove the subjective nature of the test and eradicate reader error. False-positive results can occur with myoglobinuria, oxidizing agents and peroxidases. A false-negative result can also occur with high levels of ascorbic acid, nitrite, pH < 5 and high specific gravity of the urine specimen. There is no substitute for urine microscopy performed by a competent laboratory.

Q. How would you set up a haematuria clinic?

A. Haematuria is an ideal symptom to be investigated with the one-stop clinic format. Each patient requires a full history and examination. This should include a detailed report of the haematuria, duration and any associated symptoms. Exposure to smoking and occupational history must be recorded. Examination should include an abdominal examination, external genital examination and a digital rectal examination in men. Blood tests including full blood count, urea and electrolytes, clotting screen and prostate-specific antigen (PSA) (after counselling) should be performed. An MSU and urine cytology are required. Radiological imaging of the urinary tract is required as well as flexible cystoscopy. CTU is increasingly being used as the first-line imaging modality although ultrasound can be used first-line. However, ultrasound cannot exclude the presence of upper tract tumours. Interpretation of this data can then lead to immediate discharge of the patient, or organisation of further investigations if indicated. For example in the patient with positive cytology, negative MSU, normal USS/CTU and normal urothelium on flexible cystoscopy, then diagnostic cystoscopy with random bladder biopsies, biopsy of prostatic urethra, retrograde studies and ureteroscopy are required to elucidate the cause of the high-grade malignant cells detected.

From a referral perspective, National Institute for Health and Care Excellence (NICE) Suspected Cancer: Recognition and Referral (2015) guidelines specify referral of patients who are aged 45 years and above, and have unexplained visible haematuria (VH) in the absence of urinary tract infection (UTI), or VH that persists or recurs after successful treatment of UTI. In addition, the guidelines stipulate that patients aged 60 years and over with unexplained symptomatic (e.g. associated with dysuria) non-visible haematuria (s-NVH) should be referred for investigation.

Q. How do you perform urine cytology?

A. Urine cytology is performed on a random or ideally, a mid-morning urine sample. Early morning urine samples provide degenerative specimens for cytological examination. The sample should be the whole voided urine as an MSU is the most acellular fraction. The sample needs to be rapidly transferred to the laboratory for processing. If a short delay occurs refrigeration is recommended. If longer delays are expected then prompt fixation with an equal amount of 50% alcohol can be utilised. Sensitivity of urine cytology increases with the number of specimens examined. Ideally at least three mid-morning or random specimens should be submitted for examination, but this is often impractical. Catheter specimens can be utilised but cellular changes can be seen with this method of collection. Saline washouts may also be utilised (saline barbotage) and need to be recorded on the request form. The laboratory will then centrifuge the sample, perform fixation in formalin and stain with Papanicolaou or haematoxylin and eosin dyes. The resulting slides are then analysed under the microscope for morphological changes consistent with malignancy.

Cytology is most useful in the detection of high-grade malignancy and is positive in 90% of these cases. In low-grade tumours it is positive in only 10% of patients. It is often used as a safety system for the detection of malignancy in the investigation negative group of individuals.

Q. Can you describe any alternatives to urine cytology?

A. Urine cytology has been the basis of urine testing for urothelial cancer for many decades. With a reported specificity of 95% and sensitivity of 30%-50% it is an adequate test for detecting high-risk disease but lacks the sensitivity for the detection of low-grade disease. Several novel urinary markers for urothelial carcinoma have been developed including NMP22 (nuclear matrix protein 22), BTA (bladder tumour associated antigen), BTA Stat, UroVysion and Telomerase. These urinary molecular markers generally have a high sensitivity but lower specificity compared to urine cytology. Currently, these markers are not accepted for diagnosis or follow-up in clinical guidelines. The specificity and sensitivity of these bladder cancer urinary markers are reviewed in Table 4.4.

Table 4.4 Summary of main urinary marker sensitivity and specificity in detecting urothelial carcinoma [4]

Source: Adapted from European Association of Urology. http://uroweb.org/wp-content/ uploads/EAU-Guidelines-on-Non-muscle-Invasive-BC_TaTI-20l7.pdf

NMP22 (nuclear matrix protein 22) is promoted as the ‘NMP22 BladderChek’ test. It has reported increased sensitivity compared to cytology and has been used in surveillance.

The BTA stat test identifies overproduction of the complement protective peptide complement factor H related protein (CFHrp) in the urine. This is known as the BTA. BTA TRAK is a quantitative assay of BTA.

The UroVysion test utilises fluorescent in situ hybridisation (FISH) to test for aneuploidy of chromosomes 3, 7, 17 and loss of the 9p21. This complicated test requires intact cells, expensive equipment and a dedicated laboratory.

Telomerase is overexpressed in many cancers and can be detected in the urine with the TRAP (telomerase repeat amplification) assay. Telomerase adds telomeres to chromosomal terminal DNA sequences preventing cell senescence. ImmunoCyt is a test combining cytology and immunofluorescent assay.

Q. What is the evidence for your haematuria investigation regime?

A. The National Institute for Health Research reported a lack of studies analysing the efficacy of diagnostic regimes for the investigation of haematuria [5]. Not unsurprisingly, there is wide variation in adopted haematuria investigation modalities.

However, two studies are classically quoted for the investigation of haematuria. Khadra et al. studied 1930 patients who attended a haematuria clinic from 1994 to 1997 [3]. All patients underwent a history and examination, routine blood tests, urinalysis, cytology, plain abdominal radiography, renal ultrasound, intravenous pyelogram (IVP) and flexible cystoscopy. In this cohort, 61% of patients had no pathology identified, 12% were diagnosed with bladder cancer, 13% had UTI and 2% had stone disease. The key message here is that if only ultrasound or IVP had been performed a significant proportion of upper tract malignancy would have been missed (six with ultrasound alone and at least three with IVP alone) and hence the use of both imaging modalities is recommended. Overall, macroscopic and microscopic haematuria resulted in a cancer diagnosis in 24% and 9.4%, respectively. Bladder cancer was found in more patients with microscopic haematuria than the 5% or less reported within the urological literature.

Edwards et al. studied 4020 patients attending a haematuria clinic [6]. Macroscopic haematuria resulted in a fourfold increase in the diagnosis of cancer versus microscopic haematuria (19% versus 5%). In 75% no pathology was identified. Three upper tract tumours were identified after a normal USS with 46% of this cohort receiving an IVU.

IVU has now largely been replaced by CTU given the superior sensitivity and specificity of CT over IVU. Albani et al. followed 416 patients who underwent either CTU or IVU in addition to diagnostic cystoscopy [7]. A higher sensitivity for detecting upper tract pathology was reported with CT than IVU (94.1% versus 50%). However CTU had a low sensitivity for lower tract lesions. They concluded that CTU is a more sensitive alternative to IVU in haematuria investigations but CTU does not eliminate the need for cystoscopy. Presently, there is no published large series using CTU as the main radiological investigation.

Q. Is screening for bladder cancer effective?

A. In 1992, Britton et al. investigated the use of dipstick testing for haematuria in 2356 men over 60 years old [8]. And 20% (474 patients) of these patients had non-visible haematuria on recurrent testing. Overall, 319 agreed to undergo urological examination. In this cohort 17 (5.3%) asymptomatic patients were diagnosed with bladder cancer.

In a similar study, Messing et al. screened 1575 men aged 50 years and older and found 16.4% (258 patients) had haematuria [9]. Of those with haematuria 8.1% (21 patients) were diagnosed with bladder cancer. Using the Wisconsin State Tumor Registry as an unscreened cohort for comparison, they observed a smaller proportion of muscle-invasive bladder cancer in the screened men compared to non-screened men (5%-24%). At 14 years’ follow-up, the cancer-specific mortality rate in screen-detected bladder cancer patients was zero compared to 20.4% in bladder cancer patients of the non-screened cohort.

Although these studies suggest that bladder cancer may be detected at an earlier stage through screening, there remains a high false-positive rate (85%-95%), i.e., haematuria positive but no cancer identified on investigations. Screening would There fore lead to large numbers of asymptomatic patients requiring investigation. The cost per bladder cancer diagnosis is prohibitively high at the present time.

The use of urinary dipstick testing with other urinary biomarkers, such as UroVysion, NMP22 (BladderChek) and ImmunoCyt, in bladder cancer screening in specific populations has been studied. Targeted screening of a population of smokers has reported a 3.3% pick of malignancy using cytology, dipstick and the urinary markers UroVysion and NMP22 (BladderChek) [10].

Q. If all her initial haematuria investigations are negative except for positive cytology - what now?

A. The patient with negative cystoscopy and imaging but positive cytology poses a diagnostic challenge. This patient should be offered a CTU if not already performed as part of the haematuria investigations. If the CTU is normal, then EAU guidelines [4] recommend that GA cystoscopy be performed with biopsies from normal-looking mucosa taken from trigone, bladder dome and right, left, anterior and posterior bladder walls to exclude carcinoma in situ (CIS). In men, prostatic urethra biopsies should also be taken when no tumour is seen in the bladder. Photodynamic diagnostic (PDD) cystoscopy has been shown to be more sensitive than white light cystoscopy in detecting malignant tumours, particularly CIS. If PDD equipment is available, then PDD guided bladder biopsies are recommended in positive cytology settings. Retrograde studies and ureteroscopy are also to be considered.

Q. Who should be referred to a nephrologist?

A. The Joint Consensus Statement on the Initial Assessment of Haematuria prepared on behalf of the Renal Association and British Association of Urological Surgeons (July 2008) outlined guidelines for referral to nephrology. Patients who have negative investigations from a urological perspective need nephrology referral if other factors are present:

Evidence of declining GFR by >10 mL/min within the previous 5 years or by >5 mL/min within the last 1 year

Stage 4 or 5 chronic kidney disease (eGFR <30 mL/min)

Significant proteinuria (albumin-to-creatinine ratio [ACR] of >30 mg/mmol or protein- to-creatinine ratio [PCR] of >50 mg/mmol)

Primary nephrology referral should be considered in patients under 40 years old with non visible haematuria and risk factors for severe glomerulonephritis:

Significant proteinuria (ACR of >30 mg/mmol or PCR of >50 mg/mmol)

Hypertension (BP >140/90) eGFR <60 mL/min

Note that the Joint Consensus Statement is currently under review.

Q. What are common renal (nephrological) causes of non-visible haematuria?

A. NVH from the nephrological perspective is either glomerular or non-glomerular, see Table 4.5.

Table 4.5 Nephrological disorders resulting in non-visible haematuria

Glomerular disease is a common cause of NVH. IgA nephropathy, also known as Berger’s disease, results in mesangial deposition of IgA. This is often after an upper respiratory tract infection. VH or NVH can be present. Approximately 10%-20% will develop renal failure in 10-20 years. Prognosis is worse if the patient develops hypertension or has proteinuria or fibrosis on biopsy. Treatment depends on the strict control of blood pressure and occasionally requires steroids and immunosuppression.

Alport’s syndrome is an X-linked collagen mutation resulting in blindness, deafness and nephritis resulting in chronic renal failure.

Thin basement membrane disease is an asymptomatic disease resulting in a-NVH. It is non-progressive and diagnosed only on electron microscopy of a renal biopsy.

Nephrotic syndrome results from a non-inflammatory injury to the glomerulus. Patients classically have hypoalbuminaemia, hypercholesterolaemia and hyperlipidaemia due to excessive hepatic lipoprotein synthesis. Considerable proteinuria may exist. Treatment is with steroids.

Causes of non-glomerular NVH include interstitial nephritis. This results from an inflammatory infiltrate affecting nephron function. Acute interstitial nephritis occurs 4 days to 5 weeks after starting a new drug, commonly penicillin. Symptomatically the patient develops a fever and generalised rash with oliguria, increased creatinine and hypertension.

Papillary necrosis is characterised by coagulative necrosis of the renal papillae from pyelonephritis, obstructed uropathy, sickle cell, tuberculosis, trauma, cirrhosis, analgesic nephropathy, renal vein thrombosis or diabetes (the acronym POSTCARD). Urological management involves resolution of urinary obstruction, treatment of infection and resuscitation.

A renal artery stenosis can result in renin-mediated hypertension and occurs only with a stenosis greater than 70%. Clinically a bruit may be auscultated.