Paraneoplastic glomerulonephritis

Introduction

Paraneoplastic syndrome refers to “clinical manifestations that are not directly related to tumor burden, invasion or metastasis, but are caused by secretion of tumor cell products, such as hormones, growth factors, cytokines and tumor antigens.” The idea that the kidneys could manifest paraneoplastic disease was first proposed in 1922 by Galloway who noticed the presence of an “unusual protein” in the urine of his patient afflicted with Hodgkin lymphoma. It has been proposed that the diagnosis of paraneoplastic syndrome should be suspected when: (1) there is no obvious alternate etiology for the associated syndrome; (2) a temporal relationship exists between the diagnosis of the syndrome and cancer; (3) clinical (and histologic) remission occurs after complete surgical removal of the tumor or full remission is achieved by chemotherapy; and (4) recurrence of the tumor is associated with increase of associated symptoms. Since the early 1920s, various animal studies and retrospective studies have provided evidence for paraneoplastic glomerulopathy (PNGN), but there still remains some skepticism about its existence. In addition to malignancies, benign tumors, such a pheochromocytoma, carotid body tumors, benign ovarian teratomas, and spinal cord tumors have also been reported to cause glomerular disease. This chapter will focus on the reported incidence and purported mechanisms of PNGN in solid tumors.

Epidemiology of paraneoplastic glomerulopathy in solid tumors

PNGN is likely a rare clinical entity. Less than 1% of adult cancer patients develop PNGN with overt renal disease. The true incidence of PNGN remains ill-defined for a number of reasons. First, patients with cancer can have various urinary abnormalities that are not routinely investigated, and urinalysis is not performed routinely among inpatients and outpatients with cancer. Sometimes, patients with malignancy can have subclinical glomerulonephritis (GN) manifested by presence of immune complex (IC) deposits caused by defects in the immune-regulatory mechanisms without significant renal damage. In addition, many of the reports on PNGN are based on retrospective data and case reports, which have their inherent reporting biases. The case reports of PNGN do not seem to reflect the relative incidence or prevalence of cancer in the general population. According to the American Cancer Society, lung cancer is the second most common cancer among men and women, whereas prostate is the most common cause among males and breast cancer among females. However, based on the available published reports, the solid tumors most commonly associated with a PNGN are of renal, gastrointestinal, (GI) and lung origin ( Fig. 22.1 ). Whereas solid tumors have been reported most commonly in association with membranous glomerulonephritis (MN) ( Fig. 22.2 ), a wide spectrum of glomerular lesions have been described.

Fig. 22.1, Number of reported cases of paraneoplastic glomerulonephritis in solid tumors arranged by glomerular lesion.

Fig. 22.2, Number of reported cases of various solid tumors reported with paraneoplastic glomerulonephritis.

Notwithstanding the issues of reporting bias and the relative rarity of this clinical entity, several clinical and laboratory features have been recognized to associate with paraneoplastic disease in the kidney. In a population based longitudinal study of 5425 nondiabetics without any previous diagnosis of cancer, Jorgensen et al. found that elevated albumin to creatinine ratio was associated with higher incidence of cancer, even after adjustment for age, gender, body mass index, physical activity, and smoking. Age greater than 60 years and a history of tobacco smoking has been associated with higher risk for cancer in patient with nephrotic syndrome (NS). ^,

Pathophysiology

A causal link between neoplasia and PNGN remains ill defined, but there are several purported pathogenetic mechanisms. In a study of three patients with gastric cancer presenting with NS, glomerular eluates reacted specifically with the surface of the cancer cells from the same host seen under immunofixation (IF). Thus one hypothesis is that there is deposition of circulating tumor antigen-antibody complexes in the glomeruli with subsequent activation of inflammatory pathways, which eventually produces glomerular disease. In addition, antibodies could be directed toward specific endogenous glomerular antigens. Couser et al. report a case of a patient with MN who was found to have elevated carcinoembryonic antigen following the diagnosis of NS. The patient was subsequently diagnosed with colon cancer and underwent colon resection. The authors of the report demonstrated that an antibody in the serum of the patient reacted with an antigen on the glomerular basement membrane (GBM). This antibody was removed by adsorption of serum with homogenates of the patient’s serum but not by the homogenates of normal colon, liver, or spleen.

Another possible pathogenetic mechanism is that tumors elaborate cytokines and/or permeability factors, which triggers glomerular injury. In a study on a patient with rectal adenocarcinoma associated minimal change disease (MCD), Taniguchi et al. demonstrated high levels of vascular endothelial growth factor (VEGF) in the tumor. After resection of the tumor, VEGF levels returned to normal and the proteinuria resolved. The authors hypothesized that the overexpression of VEGF by certain tumor cells was linked to the onset of proteinuria, and thus when the VEGF levels decreased after tumor resection, the proteinuria resolved. ^, ^,

Intrinsic viral oncogenic activity has been offered as an additional pathogenetic mechanism for PNGN. Viral activity may interfere with the renal clearance of the prooncogenic biological mediators released from the body. Alternatively, viral infections could lead to both malignancy and a PNGN through a common pathway. ^, The possible role of viral oncogenic activity and PNGN is discussed in more detail in the chapter on Paraneoplastic Glomerulopathy in Hematologic Malignancies.

The cellular and humoral immune systems may also have a role in the pathogenesis of PNGN. In an animal study that was performed by injecting rat colon cancer cells into immunocompetent and T-cell deficient F344 rats, the T-cell deficient rat’s kidneys did not show any morphologic abnormalities and did not have any immunoglobulin (Ig)G deposition. Thus it can be hypothesized that the presence of an intact cellular and humoral immunity is required for development of PNGN.

The various histologic patterns of PNGN may be dependent on several factors, including: (1) the duration of the neoplasm; (2) degree of tumor differentiation, which may dictate the elaboration of specific and nonspecific antigens; (3) type of tumor antigen and other products expressed; (4) type and extent of host response to these antigens via cell-mediated and antibody-mediated mechanisms; and/or (5) physiochemical characteristics of the circulating IC that render them pathogenic or nephritogenic.

Membranous nephropathy

MN is the most common PNGN associated with solid tumors. ^, Paraneoplastic MN has been reported more often in males, in those at age greater than 60 years, and in patients with a history of heavy tobacco use. ^, ^, The most commonly associated cancers are those of lung, GI, and renal origin ( Table 22.1 ). Prostate cancer is increasingly being reported with paraneoplastic MN, ^, which may reflect the increased use of prostate-specific antigen as a screening test. In most reports, patients who were diagnosed with idiopathic MN were diagnosed with a neoplasia within 12 months. However, the PNGN can present years after the neoplasia is evident.

Table 22.1

Number of Reported Cases of Paraneoplastic Membranous Nephropathy in Solid Tumors

Organ	Number of Cases
Gastrointestinal	26
Lung	27
Renal	12
Prostate	9
Breast	5

Histopathology

On light microscopy (LM), the GBMs in MN appear thickened. On electron microscopy (EM), subepithelial deposits are evident. The presence of subendothelial or mesangial deposits should raise concern for a secondary or paraneoplastic process. There are several features on histopathology that have been associated with a paraneoplastic MN. Patients with paraneoplastic MN have an increased number of infiltrating inflammatory cells in the glomeruli (> 8 cells per glomeruli). Also, whereas IgG4 is the predominant subclass found in idiopathic MN, IgG1, IgG2, and IgG3 are predominantly found in cases of secondary MN. Circulating antiphospholipase A2 receptor is seen in up to 89% cases of idiopathic membranous nephropathy. It is worth noting that circulating antiphospholipase A2 receptor is occasionally present with secondary, including paraneoplastic membranous. ^,

Pathogenetic mechanism

Heymann nephritis is an experimental rat model for active and passive immune-mediated nephritis. Megalin, which is the target antigen, localizes to the podocytes in the rat model. However, in humans, megalin is found in the proximal tubule and not in podocytes. Hence investigators have sought to find an equivalent antigen present in the podocytes to explain paraneoplastic membranous glomerulonephritis (MGN). The finding of thrombospondin type-1 domain-containing 7A (THSD7A) antibody in a patient with MGN with a neoplasm, and the disappearance of the antibody and urinary protein after the malignancy was treated, suggests a potential causal relationship between THSD7A expression and MGN. Other factors may exist, but they have yet to be fully defined. There are a number of isolated case reports in which eluates from the glomeruli of patients with known malignancy showed reactivity with the patient’s serum and the tumor homogenates, but not with normal kidney.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here