The purpose of cancer screening is to identify treatable cancers early to improve survival. Effective cancer screening therefore depends on cancer occurrence and risk, the availability and effectiveness of cancer screening tools, and expected survival in a given population. Cancer screening is beneficial if the screening tests used identify early stage, potentially curable cancers in individuals who have long expected survival in the absence of the identified cancer. Ideally, the screening test is sensitive and specific to avoid unnecessary risks and additional testing. Screening tests are also more effective if they are simple to perform and widely available. The best screening tests are also relatively inexpensive. Assuming a screening test detects a cancer at a curable stage, the effectiveness of the screening test can be assessed by calculating quality of life years saved. Implicit in this calculation is an assumption of expected survival and cancer risk in the person to be screened. Using this method, governing bodies and societies make recommendations for cancer screening in populations.
Some recommendations for cancer screening in the adult population are shown in Table 34.1 . Most recommendations account for cancer risk factors based on demographics and family history in addition to other underlying factors that are cancer and/or organ-specific. All of the recommendations assume the individual to be screened is healthy with a normal expected survival by gender and age. However, in the case of chronic kidney disease (CKD) or end-stage renal disease (ESRD), survival is lower than the normal population, and thus “routine” cancer screening recommendations require adjustment. This chapter will review the issues affecting cancer screening in those with CKD and ESRD and make recommendations for screening based on these issues.
Cancer risk in chronic kidney disease and end-stage renal disease
Many retrospective cohort studies have shown an increased incidence of cancer in ESRD patients, but the data are less clear among nondialysis CKD patients. Table 34.2 summarizes the literature examining cancer incidence among CKD patients not on dialysis. Most of these studies report hazard ratios for cancer. Although cancers of the urinary tract appear to be more common among nondialysis CKD patients, , , , , overall, the risk for most cancers does not appear to be increased in CKD patients. , , The definition of CKD varies somewhat in these studies, with some reporting specific CKD stages (see Table 34.2 ) and others relying on billing codes for diagnoses like CKD, chronic glomerulonephritis, hypertensive CKD, and diabetic kidney disease, making comparisons among studies difficult. It should be noted that the higher risk of upper urinary tract urothelial cancers is particularly prominent in studies from Asia, where these cancers also occur more frequently in the general population. The reasons for this are unclear but differences in culture, lifestyle, diet, and drug use are likely. For example, aristocholic acid present in some fruits has been identified as a carcinogenic factor and may represent a risk for this population. In addition, certain causes of CKD, like analgesic abuse (containing phenacetin) and Balkan nephropathy (also related to aristocholic acid) are also associated with an increased risk of urothelial cancers and may be more common among certain populations.
|eGFR, mL/min||Hazard Ratio Overall (CI)||Ref||Hazard Ratio Urinary Tract Cancer||Ref|
|≥ 60–75||0.98 (0.87–1.10)||4|
|≥ 45–60||0.99 (0.88–1.113)||4||1.35–1.39 (1.22–1.58)||4,5|
|< 45||1.01 (0.84–1.22)||4||1.63–1.81 (1.51–2.17)||4,5|
|< 30||1.63–2.28 (1.78–2.92)||5,10|
Unlike the CKD population, cancer incidence is usually reported to be higher among ESRD patients. – , – Table 34.3 summarizes the available literature, demonstrating higher standardized incidence ratio (SIR) among ESRD patients for many specific cancers. The SIR is the typical marker for cancer incidence and represents an estimate of the occurrence of cancer in a population relative to what might be expected if the population had the same cancer experience as a larger comparison group. In most of these observational studies, virally mediated cancers are more common in dialysis patients. These cancers include tongue and cervical cancer (associated with human papilloma virus) and liver cancers associated with hepatitis B and C. As in the nondialysis CKD group, cancers of the urinary tract are significantly more common among dialysis patients and include renal cell carcinoma (RCC), bladder cancer, and urothelial cancers of the upper urinary tract. Risk factors for these types of cancer in the kidney disease patients include acquired cystic disease in dialysis patients and analgesic abuse and Balkan nephropathy as causes of the ESRD. Most studies do not show an increased SIR for breast, lung, and colorectal cancer in ESRD patients (see Table 34.3 ). However, thyroid and other endocrine cancers appear to be more common for unclear reasons. – ,
|Bladder & ureter||1.5–16.4||1,213,14–16,17,19,20|
|Cervical & uterine||2.7–4.3||1,2|
|Thyroid & other endocrine||2.2–2.3||1,2,4,14|
The stage of cancer at the time of diagnosis is similar in ESRD patients compared with the general population except for colorectal cancers, which tend to be diagnosed earlier in dialysis patients. It is likely that gastrointestinal disorders, such as bleeding, occur commonly in dialysis patients and lead to more frequent investigative procedures that result in a cancer diagnosis. In this study, prostate cancer tended to be diagnosed later in the ESRD patients.
Cancer risk in kidney transplant recipients
Kidney transplant recipients experience an increased risk of cancer by at least 2 to 4 times compared with the age and gender-matched general population. The excess risks of cancer among transplant recipients are site specific for virally related cancers, non-melanoma skin cancer, and urothelial cancers. The greatest increased risk is seen in Kaposi sarcoma (up to 300 times), non-NMSC (2–40 times), lip cancer (more than 10 times), and cancers with a suspected viral oncogenesis, such as non-Hodgkin lymphoma/posttransplant lymphoproliferative disorder (4–16 times), and anogenital cancers. RCC is the most common urological malignancy after kidney transplantation. Compared with the general population, the risk of RCC is increased by approximately sixfold. The most common subtypes include papillary and clear cell RCC. There is a modest increased risk for other common cancer types, such as colorectal and lung cancer (by approximately 1.5–3 times), but most observational data show no increased risk for breast and prostate cancer compared with the age- and gender-matched general population.
There is now clear evidence suggesting the increased risk of cancer among transplant recipients is related to the prolonged use of immunosuppression over time. There is also emerging evidence suggesting the pattern of the excess risk may differ by the dose and types of immunosuppression. Transplant recipients who received horse antithymocyte globulin, rabbit antithymocyte globulin (thymoglobulin), or OKT3 as induction therapy had rates of lymphoma more than 20 times that of the age- and gender-matched general population. Use of alemtuzumab for induction therapy is associated with an increased risk of non-Hodgkin lymphoma, colorectal cancer, and thyroid cancer. In the context of maintenance immunosuppression, preclinical evidence has suggested that the oncogenicity of azathioprine and calcineurin inhibitors (CNIs) is associated with impairment in the ability to repair deoxyribonucleic acid (DNA) damage, thus leading to stimulation of prooncogenic cytokines and growth factors, including transforming growth factor β, vascular endothelial growth factor, and their receptors. On the contrary, trial-based evidence of mammalian target of rapamycin (mTOR) inhibitor such as sirolimus, has reported a significant reduction in the overall risk of developing skin cancer, particularly for non-melanoma skin cancer. , So far, there are three multicentered randomized controlled trials comparing the incidence of non-melanoma skin cancer as the primary outcome between mTOR inhibitors and CNIs, as maintenance immunosuppression after kidney transplantation. All found a 20% to 30% reduced risk of squamous cell carcinoma (SCC) among the mTOR inhibitor arms compared with the CNIs arm, after a follow-up time of 12 months. Also, those with a prior history of SCC experienced a longer time to the development of new SCC among the mTOR inhibitors arms.
Efficacy of cancer screening tests
Cancer screening depends on imaging, visual inspection via procedures, and laboratory and histopathologic tests, as described in Table 34.1 . The effectiveness of these in ESRD has rarely been assessed and thus the positive and negative predictive value and sensitivity and specificity of these examinations in the ESRD population are not known. Vascular calcification is common in ESRD patients and its presence may complicate interpretation of mammograms in women with ESRD. , Stool hemoccult testing is a noninvasive means of screening for colon cancer (see Table 34.1 ) but is subject to high false positive results (positive test caused by noncancer cause) in ESRD patients. Thus positive hemoccult testing in dialysis patients may contribute to higher rates of colonoscopy in this population. We do not have information on the sensitivity and specificity of fecal immunochemical or DNA testing in ESRD patients.
Tumor markers are also used as cancer screening tools but their effectiveness can be affected by kidney dysfunction, primarily caused by reduced clearance. Clearance with dialysis can also influence the use of these proteins in ESRD patients. Table 34.4 shows common tumor markers and their chemical structure and biological function that may be affected by CKD and dialysis. Most tumor markers are unreliable in ESRD patients because they are generally glycoproteins with high molecular weight, rarely removed by dialysis, and rise with hemoconcentration, yielding false-positive results in ESRD. Total prostate specific antigen (PSA) is probably valid in ESRD patients, but free PSA and free/total PSA ratios are less useful, because free PSA rises with hemoconcentration and high-ﬂux dialysis membranes affect its clearance. Controversy continues about screening for prostate cancer in the general population (see Table 34.1 ). Cancer antigen 125 (CA-125), a tumor marker for ovarian cancer, is produced by mesothelial cells and patients with any serosal ﬂuid (pleural effusion, ascites) will have elevated levels. Thus CA-125 is less useful in ESRD patients, particularly those on peritoneal dialysis. β-human chorionic gonadotropin and α-fetoprotein are probably reliable in ESRD patients.