Introduction
Paraneoplastic syndromes refer to manifestations of cancer that are not related to tumor burden, invasion, or metastatic disease. The manifestations may be systemic or organ-limited and can involve virtually any organ system. There is a paucity of literature on paraneoplastic kidney disease (PnKD), likely because of the rarity of these conditions. This narrative review will focus on PnKDs that occur in the setting of solid tumors and hematologic malignancies.
Because most of the literature on the epidemiology of PnKD comprises case reports and case series, the incidence and prevalence of PnKD are difficult to ascertain. Some data have shown that the risk for cancer is higher in patients with glomerular disease than in the general population. For example, the Danish Kidney Biopsy Registry, which includes all kidney biopsies performed in Denmark since 1985, reported that the risk for cancer at 1 year and 1 to 4 years after the diagnosis of glomerulopathy was increased by 2.4- and 3.5-fold, respectively, compared with the general population. In a population-based study in Tromsø, Norway (originally designed to evaluate cardiovascular disease), the albumin-creatinine ratio at baseline for patients with glomerular disease significantly correlated with the incidence of cancer during 10 years of follow-up. Further, those with albumin-creatinine ratios in the highest quintile were 8.3- and 2.4-fold more likely to be diagnosed with bladder cancer and lung cancer compared with the lowest quintile. ,
The pathophysiology of PnKD is usually linked to products of tumor cells or from paraneoplastic autoimmune manifestations that cause kidney damage. , This mechanism is well-characterized in certain paraneoplastic neurologic syndromes. For example, anti-Hu (ANNA1) antibody in the setting of small cell lung cancer is associated with encephalomyelitis, sensory neuropathy, and paraneoplastic cerebellar degeneration, and anti-Yo (PCA1) antibody in ovarian and breast cancers is associated with paraneoplastic cerebellar degeneration. However, with the exception of monoclonal gammopathies, the exact pathomechanisms linking tumor products and renal disease are poorly understood.
In patients diagnosed with PnKD, the temporal profile of the cancer diagnosis and kidney manifestations may be asynchronous. It is generally accepted that malignancies are present for up to several years before becoming clinically apparent. Patrone et al. applied a modified version of Collins’ law, which was originally used to estimate the time-to-recurrence of Pediatric Wilms tumors, in order to estimate the time-to-recurrence of adult solid tumors. Further, if more than 6 years have elapsed since remission of the cancer, it is unlikely that a glomerulopathy is related to the malignancy. To illustrate this point further, in the study by Lefaucher et al., the tumor was clinically evident in only 52% of the patients, underscoring the fact that cancers may be clinically silent and unless looked for systemically, may lead to unnecessary treatment of membranous nephropathy (MN).
The lack of published criteria for PnKD makes it challenging to distinguish between kidney disease occurring coincidentally in a patient with cancer and true PnKD. Box 21.1 summarizes “criteria” that have been suggested to help strengthen the link between cancer and kidney disease, when suspected. There are several case reports that satisfy criteria 1 and 2, but criteria 3 is generally lacking. The exceptions are plasma cell dyscrasias and B-cell disorders, which produce monoclonal antibodies that are directly toxic to the kidney.
- 1.
A renal syndrome and cancer that develop concurrently or within a few years of each other
- 2.
The renal syndrome resolves or improves significantly after cancer treatment without immunotherapy and relapses with recurrence of cancer
- 3.
A pathophysiologic link is present between the cancer and kidney disease
Among the PnKD, and despite not meeting criteria 3, solid tumor-associated MN and Hodgkin lymphoma-associated minimal change disease (MCD) have become recognized as typical forms of paraneoplastic glomerulonephritides. Other glomerular diseases mentioned in this chapter are rarely associated with cancer, with the association between the kidney lesions and cancer infrequently fulfilling the criteria listed.
Finally, it is important to note that treatment paradigms for PnKD may be quite different from noncancer-related glomerular disease, in that the main strategy in PnKD is treatment of the underlying cancer. Patients with Hodgkin disease-associated podocytopathy and solid cancer-associated MN are frequently resistant to immunosuppressive regimens that are traditionally used in these entities but respond well to treatment of the cancer.
Paraneoplastic kidney diseases in patients with hematologic malignancies
Paraprotein-associated disorders are caused by clonal plasma or B-cell populations that produce monoclonal immunoglobulins (Igs), which may cause kidney damage. These antibodies have physiochemical properties that lead to specific types of kidney injury, leading to a variety of histologic patterns on kidney biopsy ( Box 21.2 ). An important point to emphasize is that the presence of end-organ damage may prompt treatment even if the underlying clone does not officially meet criteria for cancer. This necessitates familiarity with the criteria for overt multiple myeloma, smoldering multiple myeloma, and monoclonal gammopathy of undetermined significance (MGUS) ( Box 21.3 ). For example, the majority of patients with AL (light chain) amyloidosis have >10% clonal plasma cells found on bone marrow biopsy, and thus do not meet criteria for multiple myeloma. Nonetheless, the kidneys are commonly affected by AL amyloidosis, and the disease itself is associated with significant morbidity and mortality. Similarly, in other paraprotein diseases of the kidney, the presence and detection of renal involvement may be the factor that prompts treatment of the underlying clone. In this section, we will highlight selected paraprotein-mediated disorders. Many of these are discussed in Chapter 6, Chapter 7, Chapter 8, Chapter 9 .
Myeloma (light chain) cast nephropathy
Monoclonal immunoglobulin deposition disease
Light chain deposition disease
Heavy chain deposition disease
Heavy and light chain deposition disease
Light chain proximal tubulopathy
Type I cryoglobulinemic glomerulonephritis
Paraprotein-associated C3 glomerulonephritis
Proliferative glomerulonephritis with monoclonal immunoglobulin deposits
Monoclonal immunoglobulin (AL/light chain, AH/heavy chain, ALH/light and heavy chain) amyloidosis
Immunotactoid glomerulopathy
Multiple myeloma
Clonal plasma cells in the bone marrow ≥ 10%, or biopsy-proven plasmacytoma, and any myeloma-defining event (hypercalcemia, renal insufficiency, anemia, bone lesions)
Any of the following:
- ■
Clonal plasma cells in the bone marrow ≥ 60%
- ■
Serum free light chain ratio of ≥ 100 (involved/uninvolved)
- ■
More than 1 focal lesion on MRI
- ■
Smoldering multiple myeloma
Serum M protein (IgG or IgA) ≥ 30 g/L or urine M protein ≥ 500 mg/24 hours and/or 10%–60% clonal plasma cells in the bone marrow AND no evidence of myeloma defining events or amyloidosis
Monoclonal gammopathy of undetermined significance
Presence of monoclonal gammopathy:
- ■
serum concentration of IgM or non-IgM monoclonal protein < 30 g/L
- ■
abnormal serum free light chain ratio (with increased level of involved light chain)
- ■
urinary monoclonal protein < 500 mg/24 hours
- ■
< 10% clonal plasma cells on bone marrow biopsy
No evidence of myeloma defining event, amyloidosis, or systemic lymphoma
Ig , Immunoglobulin; M , monoclonal; MRI , magnetic resonance imaging.
Myeloma cast nephropathy
Although not a glomerular disease, myeloma cast nephropathy (MCN), also known as myeloma kidney , is a prototypical paraneoplastic renal disease. The pathogenesis of cast nephropathy is linked to the binding of Ig free light chains to uromodulin (Tamm-Horsfall protein), causing tubulointerstitial injury via precipitation of light chain casts in the distal nephron. This occurs via interaction or via the binding of free light chain complementary determining region-3 with the light chain binding domain of uromodulin. , The incidence of MCN is not known because the vast majority of patients with myeloma do not undergo a kidney biopsy. Severe renal failure (often requiring dialysis) is common in patients presenting with biopsy-proven MCN, and compared with other kidney diseases associated with monoclonal gammopathies, the diagnosis of MCN is supported if only a small percentage of total proteinuria is composed of albumin. Having MCN on kidney biopsy portends a worse overall prognosis than having light chain deposition alone.
The first principle of treating MCN is to rapidly lower serum free light chain levels by targeting the underlying plasma cell clone, which is associated with improved renal outcomes. Antiplasma cell strategies have expanded greatly in the last 2 decades, including the proteasome inhibitor bortezomib (US Food and Drug Administration– approved for the treatment of multiple myeloma in 2003) and high-dose melphalan/autologous stem cell transplantation. Patients achieving renal response with treatment have improved overall and renal survival ( Table 21.1 ).
| Renal Response | eGFR at Baseline (mL/min/1.73m 2 ) | Best Creatinine Clearance |
|---|---|---|
| Complete response | < 50 | > 60 mL/min |
| Partial response | < 15 | 30–59 mL/min |
| Minor response | < 1515–29 | 15–29 mL/min30–59 mL/min |
The use of adjunctive, extracorporeal therapies for additional light chain removal is controversial. Plasma exchange therapy has been used with mixed results in randomized controlled trials and retrospective case series. The interpretation of these data is complicated by small patient numbers, the small percentage of patients undergoing a kidney biopsy to confirm the diagnosis of MCN, and their publication before the era of modern antiplasma cell therapies.
High cut-off hemodialysis has been advocated in MCN with the theoretical benefit of removing pathogenic light chains from the circulation. A recently published study randomized 98 patients in France with renal failure requiring dialysis and biopsy-proven MCN to treatment with intensive high cut-off hemodialysis versus conventional dialysis, in addition to standardized chemotherapy (bortezomib with dexamethasone). Dialysis independence was not different between treatment arms at 3 months (primary endpoint, 41% vs. 33%, p =.42), although more patients treated with high cut-off dialysis were dialysis-free at 12 months (secondary endpoint, 61% vs. 38%, p =.02). The EuLite (EUropean trial of free LIght chain removal by exTEnded haemodialysis in cast nephropathy) study is another randomized controlled trial studying high cut-off dialysis in MCN, the results of which are forthcoming (clinicaltrials.gov ID NCT00700531).
There has also been interest in treating cast nephropathy by inhibiting the interaction of free light chains with uromodulin in the nephron. A competitive inhibitor of this interaction prevents renal failure in a rat model of MCN. No studies have been published to date exploring this approach in humans.
Light chain (AL) amyloidosis
Light chain (AL) amyloidosis is a paraneoplastic disease where within an amyloidogenic monoclonal Ig light chains deposit organs and cause damage. Heavy chain (AH) and light/heavy chain (ALH) amyloidosis occur when the amyloidogenic protein is composed of heavy or light + heavy chains, respectively. These are much less common and are approached similarly to AL amyloidosis, and will not be discussed further. The majority of cases of AL amyloidosis are caused by plasma cell clones that do not meet criteria for multiple myeloma, with B or lymphoplasmacytic clones being rare. The kidney is one of the most common organs affected in AL amyloidosis, with up to 70% of patients demonstrating renal involvement, and which presents with proteinuria (commonly nephrotic syndrome) and renal insufficiency. It should also be noted that patients with vascular and interstitial renal AL amyloidosis (i.e., without glomerular involvement) can present with minimal proteinuria.
Historically, the prognosis for patients with AL amyloidosis has been extremely poor. The median survival after diagnosis was 7 months until the 1990s, when treatment with melphalan and prednisone was shown to extend patient survival. Subsequently, the use of high-dose melphalan followed by autologous stem cell transplantation, and bortezomib-based chemotherapeutic regimens, have dramatically improved patient outcomes. The presence and degree of cardiac involvement are the main determinants of patient survival, and cardiac response (changes in N-terminal pro b-type natriuretic peptide levels) with treatment are associated with survival. Baseline renal parameters (estimated glomerular filtration rate [eGFR] and proteinuria) and their response to therapy may be predictive of renal survival in AL amyloidosis. Renal progression (≥ 25% decrease in eGFR) may be associated with worse renal prognosis, whereas renal response (> 30% decrease in proteinuria to < 0.5 g/day and nonprogression of eGFR) is associated with improved renal outcomes. Recent data suggest that patients with end-stage renal disease may benefit from treatment with high-dose melphalan followed by autologous stem cell transplantation, which then may allow them to undergo successful kidney transplantation.
The renal response to therapy may be much slower than that observed for other glomerular diseases, possibly because the amyloid that has deposited in the kidneys may persist for years after treatment. Emerging adjunctive therapies include antiamyloid antibodies to improve organ function by removing amyloid deposits. ,
The monoclonal immunoglobulin deposition diseases
The monoclonal immunoglobulin deposition diseases (MIDDs) comprise light chain deposition disease (LCDD), heavy chain deposition disease (HCDD), and light and heavy chain deposition disease (LHCDD). LCDD is the most common of the MIDDs, and the majority of cases occur in the setting of multiple myeloma or nonmyeloma plasma cell clones. On kidney biopsy, immunofluorescence microscopy shows staining for kappa, or less commonly lambda, light chain, without staining for Ig heavy chain. Although glomerular involvement is common, the pathognomonic finding on kidney biopsy is the presence of light chain deposits in the tubular basement membranes. Extrarenal involvement may occur, with one recent series of patients with multiple myeloma and LCDD finding cardiac involvement in one-third of patients.
HCDD is much less common than LCDD and is characterized by immunofluorescence staining for Ig heavy chain without light chain on kidney biopsy. IgG is the most common involved heavy chain with IgA HCDD being less common. The molecular characterization of HCDD is based on a deletion in the heavy chain constant domain 1 (CH1) region of the Ig, which results in the production and secretion of a truncated heavy chain by plasma cells that in turn deposits in the kidney. Indeed, virtually all patients with HCDD have this mutation, and HCDD can be reproduced in a mouse model by introducing the CH1 mutation. Antiplasma cell therapy, including bortezomib-based regimens and high-dose melphalan/autologous stem cell transplantation, results in hematologic response rates in the majority of patients, and renal response is predicated on attaining hematologic response. ,
Lymphoplasmacytic lymphoma/waldenström macroglobulinemia
Lymphoplasmacytic lymphoma, also known as Waldenström macroglobulinemia , results in the production of IgM proteins that can cause end-organ damage. Renal disease is rare, with a recent case series describing 3% of patients with lymphoplasmacytic lymphoma developing kidney disease. However, the presence of monoclonal IgM deposition on kidney biopsy should raise suspicion for lymphoplasmacytic lymphomas, because a recent retrospective series found that lymphoplasmacytic lymphoma was present in (74%) of patients with IgM-mediated kidney disease. The IgM paraprotein in lymphoplasmacytic lymphoma may exhibit amyloidogenic or cryoglobulinemic properties, with kidney biopsy findings, and a variety of histologies can be observed on kidney biopsy (amyloidosis, infiltration by the lymphoplasmacytic lymphoma, LCDD, light chain cast nephropathy). As with all paraprotein-mediated kidney diseases, management of monoclonal IgM-mediated kidney disease focuses on treatment of the underlying lymphoplasmacytic clone.
Type I cryoglobulinemic glomerulonephritis
Cryoglobulins are Igs that have the physiochemical property of precipitating at cold temperatures. Type I cryoglobulins are monoclonal Igs that have the properties of cryoglobulins, and these proteins are produced by underlying B- or plasma cell clones. Recent case series have described type I cryoglobulins comprising 10% to 22% of all detectable cryoglobulins, with the remaining containing mixed (including polyclonal) Igs. More than half of patients have cutaneous involvement, and renal disease occurs in 14% to 20% of patients, with glomerulonephritis being the most common finding on kidney biopsy. ,
The kidney biopsies in patients with type I cryoglobulinemic renal disease usually show endocapillary or membranoproliferative glomerulonephritis on light microscopy. Cryoplugs (periodic acid–Schiff-positive, microvascular thrombi) and/or overt vasculitis may be present in some cases. Congo red staining is negative except in rare cases where the cryoglobulin also has amyloidogenic properties. Immunofluorescence microscopy exhibits staining for the involved monoclonal Ig. On electron microscopy, the cryoglobulins manifest as electron dense deposits in a subendothelial and mesangial distribution. These deposits may exhibit organized substructure in the form of large fibrils, tactoids, and microtubules. Management of type I cryoglobulinemic glomerulonephritis focuses on diagnosis and treatment of the underlying clonal cell disorder. Plasma exchange therapy has been used in cases of severe, life-threatening vasculitis and rapidly progressive glomerulonephritis.
Monoclonal gammopathies of renal significance
The monoclonal gammopathies of renal significance (MGRS) do not meet criteria for multiple myeloma or systemic lymphoma. Histologically, these can result in many of the biopsy findings found in Box 21.2 . These renal pathologies can be found in patients with myeloma or lymphoma, but the majority of cases do not meet criteria for malignancy. , , Similar to multiple myeloma and AL amyloidosis, the diagnosis and treatment focuses on the underlying clonal cell disorder, and recent data support that achieving a hematologic response is associated with renal outcomes in these patients.
A specific challenge exists in MGRS cases where the workup does not reveal evidence of an underlying plasma or B-cell clone. This is the case for the majority of patients with proliferative glomerulonephritis with mononuclear immunoglobulin deposits, and these patients also commonly do not have a detectable paraprotein in the blood or urine. One hypothesis to explain this discrepancy is that these patients have low level clones and/or paraproteinemia that are not detectable by current screening methods, but this remains to be proven. Recent uncontrolled data suggest that these patients’ renal outcomes may improve with empiric, chemotherapy-type regimens that target a hypothesized underlying clone. These observations require reproduction and validation, particularly given that most nephrologists do not have access to these drugs, and they may cause severe adverse events.
Membranous nephropathy
The prevalence of cancer in patients with MN is between 6% and 22%. , , The risk of cancer in patients with MN increases with age. Patients exhibit symptoms of malignancy in about half of cases, with the remainder being asymptomatic. Cancers of the lung and prostate are the malignancies most frequently associated with MN. There are no clinical features on presentation that distinguish cancer-associated MN from idiopathic MN (iMN).
Histologically, the light (glomerular basement membrane thickening) and immunofluorescence microscopy (granular staining for IgG, kappa, lambda and C3 in a subepithelial distribution) findings are indistinguishable from primary MN ( Fig. 21.1 A and B ). However, there may be other histologic features that distinguish cancer-associated MN from iMN. One study found that having more than eight inflammatory cells per glomerulus was associated with cancer-associated MN versus iMN (sensitivity 92% and specificity 75%). Analysis of the IgG subclasses of the immune deposits in cancer-associated MN shows a predominance of IgG1 and IgG2, rather than IgG4 predominance that is common in idiopathic MN. Antibodies to the transmembrane glycoprotein M-type phospholipase A2 receptor (PLA2R) is typical of primary MN and is now thought to distinguish this autoimmune entity from secondary MN (e.g., in lupus). However, in one study, serum anti-PLA2R antibodies were found to be present in three of 10 patients with MN in whom cancers were subsequently detected. The fact that anti-PLA2R antibodies persisted despite resection of the tumor and that IgG4 was the predominant Ig subclass suggests that these cases were likely coincidental primary MN rather than cancer-associated MN.
