Abstract
Background: Membranous glomerulonephritis (MGN) can be found in patients with cancer as a paraneoplastic syndrome or it could be manifested clinically before tumor detection. The aim of this study was to evaluate the frequency and type of renal histopathological alterations in patients with malignancy that died without cancer treatment and were submitted to necropsy. Methods: Patient's demographical and clinical data collection and laboratory tests (serum creatinine and urine sample) were evaluated. Results: Kidney fragments from 21 patients were obtained and studied by light microscopy after habitual staining. Immunohistochemistry studies were performed with monoclonal immunoglobulin and tumor markers. Patients' mean age was 71 years and 62% were male. The most frequent tumor was gastric cancer (five cases), followed by colon and oral cavity (three cases each). In 67% of the cases, malignancy was the main cause of death. Serum creatinine was increased in 10 cases, proteinuria in 15, and hematuria was present in 8 cases. The most usual glomerular lesion found was thickening of basement membrane (BM) of the capillary loops. There were two cases of IgA nephropathy, three cases of focal segmental glomerulosclerosis, and one case of MGN. Only in the patient with MGN and metastatic melanoma specific tumor markers were identified in the kidney. Conclusions: We observed a wide range of glomerular pathological changes and abnormal urinary sediments in almost all patients, but we found tumor marker deposits only in the patient with MGN.
INTRODUCTION
Glomerulopathy is a morphological and functional renal disorder that may be found in patients with cancer as a paraneoplastic syndrome or may be clinically expressed before tumor emergence.Citation1—5
Membranous glomerulonephritis (MGN) is a condition characterized by the thickening of the membrane of glomerular capillary loops and is the most frequent cause of nephrotic syndrome in adults.Citation6—8 Immunopathological and electron microscopy studies in these patients suggest an immune complex pathogenesis for this disease.Citation9 Secondary MGN may be seen in the setting of autoimmune diseases, infections and neoplasias, and with some therapeutic agents.Citation2,Citation10 In MGN cases, the clinical manifestations may precede the malignancy or occur simultaneously with it.Citation8,Citation9,Citation11
The prevalence of renal involvement in cancer patients has been analyzed in autopsy and clinical studies. In these studies, mesangial and sub-endothelial immunoglobulin deposits were observed quite often (15–50%), whereas subepithelial deposits have not been demonstrated.Citation9,Citation11,Citation12 This association has been reported with lung bronchial, esophageal, gastric, colonic, renal, and breast carcinomas and melanomas,Citation11,Citation13—17 and more rarely with ovary, bladder, prostate, uterine cervix, bile ducts, thyroid, breast, and pancreas carcinomas, lymphomas, neuroblastomas, leukemia, lipomas, orbital malignant tumors, and malignant pleural mesotheliomas.Citation3,Citation18—23 In addition, some studies suggested that the tumor's recurrence is often accompanied by proteinuria and clinical resolution or recurrence of proteinuria is directly correlated with tumor activity.Citation5,Citation8,Citation11,Citation12,Citation17 Although a temporal relationship between these two events has been suggested, the pathophysiological mechanisms involved in these cases are not clear.Citation5,Citation8,Citation9,Citation11—13,Citation15,Citation17,Citation22—24 The aim of this study was to assess the frequency of glomerular disorders in patients dying with malignancy and submitted to autopsy.
PATIENTS AND METHODS
After approval by the Ethics Committee of our Institution, 1004 patients submitted to necropsy in the Pathology Department from January 2002 to December 2003 were analyzed. Among them, 71 (7.07%) had cancer diagnosis. From these 71 patients, 21 were selected with histological diagnosis of malignancy and no previous oncological treatment (). The exclusion criteria were patients younger than 18 years and history of nephropathy due to infection, systemic lupus erythematosus, or liver disease. Patient's data collection and laboratory exams (serum creatinine and urine sample) in the last 6 months preceding the diagnosis of cancer were evaluated. There were assessed three laboratory tests for serum creatinine and urine, and the presence of arterial hypertension and/or diabetes. Serum creatinine was considered normal when it was ≤1.3 mg/dL for men and ≤1.1 mg/dL for women.
TABLE 1. Clinical, laboratorial, and histological diagnosis of the patients
In each autopsy, macroscopic and microscopic examination was performed and two renal parenchyma fragments were processed and stained for hematoxylin and eosin (HE), periodic acid methenamine silver (PAMS), and Congo red and submitted to immunohistochemical study using the monoclonal antibodies IgA (The Binding Site, Birmingham, UK, 1:300), IgG (The Binding Site, 1:400), IgM (The Binding Site, 1:100), and C9 (The Binding Site, 1:400). Tumor antibodies in kidney were studied using immunohistochemical techniques for high- and low-molecular-weight cytokeratin (AE1 and AE3, DAKO, California, USA, 1:200) in cases of carcinomas and using HMB-45 (Novocastra, Wetzlar, Germany, 1:300) for melanomas, according to the primary tumor histological diagnosis. Renal parenchyma was examined by light microscopy, using a previously established protocol.Citation25
RESULTS
The mean age for these 21 patients was 71 years (ranging from 44 to 101 years and median 74 years), and 13 (62%) of them were male. The higher frequency of tumor occurred in stomach (five cases), followed by colon (three cases), oral cavity/pharynx (three cases), esophagus (two cases), breast (two cases), prostate (two cases), unknown origin (one case), uterine cervix (one case), melanoma (one case), and pancreas (one case).
We had to exclude 50 (70.4%) patients due to the following criteria: previous chemotherapy for cancer (56%), no laboratory test information in the last 6 months preceding the diagnosis of cancer (42% of the cases), and untreated hepatitis C (2%). These exclusions were performed because patients who received previous chemotherapy for cancer and patients with evidence of any liver disease (hepatitis C) may have renal glomerular alterations.
Cancer was the main cause of death, as described by the death certificate, in 67% of these 21 cases. In seven patients (33%) other causes of death were identified, such as bronchopneumonia (four cases), acute myocardial infarction, heart failure, acute cholecystitis, and chronic obstructive pulmonary disease (one case each).
Serum creatinine level was increased in 10 patients. In only one patient there were no alterations in the urine sample (hematuria or proteinuria). Proteinuria was identified in 15 (71.4%) cases, 8 patients also had hematuria, and in 5 patients only hematuria was observed (23.8%). No patient had history of surgery in the past 3 years of the cancer diagnosis. Hypertension was identified in four patients who were on antihypertensive drugs.
The Spearman analysis indicated that cancer histological diagnosis and proteinuria was positively correlated (rs = +0.176, p = 0.444) and positive correlation noted between the histology and hematuria (rs = +0.232, p = 0.311). The p-value of <0.05 was considered to be statistically significant.
The histological examination (HE) identified 11 cases of thickening of the glomerular basement membrane (BM), 4 cases with expansion of mesangial matrix, 4 cases with mesangial hypercellularity, and only 5 cases with normal glomerulus. Immunohistochemical study was positive for IgM in six patients (28.7%), for IgA in three (14.3%), for IgG in three (14.3%), and for C9 in six (28.6%). In some cases, more than one marker was positive (see ). Eleven cases (52%) had no immunohistochemical expression of these markers.
TABLE 2. Renal findings and correlation with cancer
The renal histology diagnosis was focal segmental glomerulosclerosis in three cases, IgA nephropathy in two cases, and MGN in one case. This last case (MGN) showed positive immunohistochemistry expression for IgA, IgG, IgM, and C9 positive in addition to positive expression of HMB-45 and S-100, which are markers for melanoma. In the other cases of carcinoma, there was no renal cytokeratin immunohistochemical positivity. In all the cases that showed renal tissue positivity for some immunoglobulin or complement, the patients also had proteinuria and/or hematuria.
DISCUSSION
In this study, renal laboratory changes such as proteinuria, hematuria and creatinine increase, renal tissue immunoglobulin and complement deposition, and glomerular structural abnormalities (thickening of the glomerular BM, expansion of mesangial matrix, mesangial hypercellularity, focal segmental glomerulosclerosis, IgA nephropathy, and MGN) were found in almost all patients studied, even considering that none of them had previous diagnosis of renal disease.
Many authors described that clinical manifestations of MGN may precede the malignancy in 40% of the cases or present simultaneously with cancer diagnosis in 80% of the cases.Citation9,Citation11,Citation24,Citation26—28 In fact, in the present series, one patient was diagnosed with MGN 6 months before a melanoma was detected. This event was also described previously in other three patients, who presented MGN as the first sign of an occult melanoma.Citation11,Citation13,Citation29 In the present case, to the best of our knowledge for the first time in the literature, the presence of tumor markers (HBM45 and S-100 antigens) was detected in the glomerulus. On the other hand, using high- and low-weight cytokeratin marker AE1+AE3, which is a pool of different cytokeratins antibodies that can detect almost all types of carcinoma, we were unable to find any positivity in the other patients with different types of glomerular injury. This discrepancy might be related to different pathogenic mechanisms. Rihova,Citation5 reported that malignant tumors associated MGN, might be caused by the production of antibodies against tumor cells deposition immune complexes and glomerular deposits. In the same way, Lefaucheur,Citation8 hypothesized that tumor-related glomerular injury might be related to the inflammatory cells evoked by the tumor.
In this study, 95.2% of the patients presented proteinuria and/or hematuria before cancer diagnosis. According to LefaucheurCitation8 and Burstein,Citation9 the presence of proteinuria before cancer diagnosis and tumor recurrence correlates directly with tumor activity and could be used as a follow-up exam. It is interesting to point that 42.8% of the patients presented renal immunoglobulin deposition when renal tissue was examined in the autopsy.
A positive correlation was also found between the cancer diagnosis histology (adenocarcinoma and squamous cell carcinoma and melanoma) and proteinuria and hematuria. Our study revealed no significant association of cancer diagnosis histology and hematuria and proteinuria. This result is in accordance with the findings reported by other authors.Citation5,Citation8 These studies have also shown that the histological type of cancer and proteinuria may have an important role in activate circulating tumor cells.
The fact that the mean age of our cancer patients was 71 years and elderly people have more chance to have any kidney injury could question the possibility of the association of renal lesion and cancer to be just by chance. However, the incidence of glomerulopathy in our patients was higher (28.6%) than the normal old population.
Recent studies have suggested that glomerulonephritis secondary to other causes increases in patients below 16 years and above 60 years.Citation1,Citation2,Citation10 The fact that the mean age of our cancer patients was 71 years and elderly people have more chance to have any kidney injury, could question the possibility of the association of renal lesion and cancer be just by chance. However, the incidence of glomerulopathy in our patients was very high (28.6%) than the normal old population. Studies show that due to the association of cancer with MGN and nephrotic syndrome, if patient is above 50 years of age and presents a history or evidence of renal clinical manifestation, an occult cancer should be investigated.Citation8,Citation13
This study has important clinical implications. Other glomerulopathies besides MGN were found in association with cancer in our patients. In two of them, IgA nephropathy was diagnosed. In these patients, there was no history of hypertension, but both showed glomerular mesangial hypercellularity, deposition of IgA, proteinuria, and hematuria. There are no previous studies relating the occurrence of IgA nephropathy with cancer. Focal segmental glomerulosclerosis was identified in three cases, all with proteinuria and hematuria. However, none showed abnormal creatinine, suggesting that this glomerular injury might be related to the tumor.
The treatment of cancer often requires chemotherapy and in that case it is very important to evaluate the presence of renal disease. Early diagnosis of renal disease could have significant impact on the patients' prognosis and might be possible to prevent worsening of renal injury due to the cancer treatment.
In summary, this study identified glomerular immunoglobulin deposits and the presence of abnormal urinary sediments in almost all patients studied. We also observed a wide range of glomerular pathological changes in these patients, but we found tumor markers deposits only in the patient with MGN.
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
Related Research Data
REFERENCES
- Glassock RJ. Secondary membranous glomerulonephritis. Nephrol Dial Transplant. 1992;7(suppl 1):S64–S71.
- Jefferson JA, Couser WG. Therapy of membranous nephropathy associated with malignancy and secondary causes. Semin Nephrol (Review). 2003;23:400–405.
- Meydanli MM, Erguvan R, Altinok MT, Ataoglu O, Kafkasli A. Rare case of neuroendocrine small cell carcinoma of the endometrium with paraneoplastic membranous glomerulonephritis. Tumori. 2003;89:213–217.
- Yoshimoto K, Yokoyama H, Wada T, Pathologic findings of initial biopsies reflect the outcomes on membranous nephropathy. Kidney Int. 2004;65:148–153.
- Rihova Z, Honsova E, Merta M, Secondary membranous nephropathy – once center experience. Renal Failure. 2000;27(4):397–402.
- Zucchelli IP, Cagnoli L, Pasquali C. Clinical and morphologic evolution of idiopathic membranous nephropathy. Clin Nephrol. 1986;25:282–288.
- Wasserstein AG. Membranous glomerulonephritis. J Am Soc Nephrol. 1997;8:664–668.
- Lefaucheur C, Stengel B, Nochy D, Membranous nephropathy and cancer: Epidemiologic evidence and determinants of high-risk cancer association. Kidney Int. 2006;70:1510–1517.
- Burstein DM, Korbet SM, Schwartz MM. Membranous glomerulonephritis and malignancy. Am J Kidney. 1993;22:5–10.
- Blumberg A, Steanm B, Wegnann W, Schaub N. Immune-mediated type a gastritis and glomerulonephritis. Am J Kidney Dis. 1993;21:210–212.
- Luyckx C, Vandamme B, Vanrenterghem Y, Maes B. Carcinoid tumor and membranous glomerulonephritis: Coincidence or malignancy-associated glomerulonephritis? Clin Nephrol. 2001;57:80–84.
- Ohtani H, Wakui H, Komatsuda A, Distribution of glomerular IgG subclass deposits in malignancy-associated membranous nephropathy. Nephrol Dial Transplant. 2004;19:574–579.
- Gomez CL, Prados C, Alvarez-Sala R. Membranous glomerulonephritis and melanoma: A causal correlation? J Nat Câncer Inst. 1994;86:64–65.
- Shikata Y, Hayashi Y, Yamazaki H, Shikata K, Makino H. Effectiveness of radiation therapy in nephrotic syndrome associated with advanced lung cancer. Nephron 1999;83:160–164.
- Ashman N, Steele JPC, Sheaff M, Rudd RM, Onwubalili J. Membranous nephropathy resolving with treatment of bronchial carcinoid tumor. Am J Kidney Dis. 2000;36:E15.
- Grivaux M, Renault D, Gallois JC, Barbanel C, Blanchon F. Small-cell lung cancer disclosed by extramembranous glomerulonephritis. Rev Mal Respir. 2001;18:197–199.
- Kijima Y, Yoshinaka H, Owaki T, Breast cancer with nephritic syndrome: Report of two case. Surg Today. 2004;34:755–759.
- Porush JG. Membranous nephropathy and prostatic carcinoma. Nephron 2000;86:536.
- Matsuura H, Sakurai M, Arima K. Nephrotic syndrome due to membranous nephropathy associated with metastatic prostate cancer: Rapid remission after initial endocrine therapy. Nephron 2000;84:75–78.
- Lai KN, Li PK, Lui SF, Membranous nephropathy related to hepatitis B virus in adults. N Engl J Med. 1991;23:3241457.
- Tourneur F, Bouvier R, Langue J, Membranous nephropathy and orbital malignant tumor. Pediatr Nephrol. 2000;14:53–55.
- Sakamoto K, Suzuki H, Jojima T. Membranous glomerulonephritis associated with diffuse malignant pleural mesothelioma: Report of a case. Surg Today. 2000;30:1124–1126.
- Fang JT, Chen YC, Hsueh S. Lymphatic tumor emboli of perirenal fat in patient with nephrotic syndrome receiving renal biopsy, ultimately revealed gastric adenocarcinoma with membranous glomerulonephritis. Renal Failure. 2001;23:743–748.
- Birkeland SA, Storm HH. Glomerulonephritis and malignancy: A population-based analysis. Kidney Int. 2003;63:716–721.
- Jennette JC, Falk RJ. Nephritic syndrome and glomerulonephritis. In: Silva FG, D'Agati VD, Nadasdy T, eds. Renal Biopsy Interpretation. Churchill Livingstone: New York; 1996.
- Lee JC, Yamauchi H, Hopper J. The association of cancer and the nephrotic syndrome. Ann Intern Med. 1966;64:41–51.
- Costanza ME, Pinn V, Schwarz RS, Nathanson L. Carcinoembryonic antigen–antibody complexes in a patient with colonic carcinoma and nephrotic syndrome. N Engl J Med. 1973;289:520–522.
- Eagen JW. Glomerulopathies of neoplasia. Kidney Int. 1977;11:297–303.
- Baris YS, Akpolat T, Akpolat I. Coexistence of membranous glomerulonephritis and Kaposi sarcoma. Nephron. 1998;79:371–372.