Pathological, Clinical Characteristics and Prognostic Analysis of Amyloid Nephropathy


 Background: The number of patients diagnosed with renal amyloidosis is increasing year by year, and the pathological grade and prognostic factors are not clear. This study attempts to investigate the pathological features of renal amyloidosis and the prognostic factors in patients with amyloidosis. Method：To evaluate the pathological sections of 184 patients with renal amyloidosis in the Department of Pathology, Shandong University, and to retrospectively analyze the clinical data of 175 patients with amyloidosis in our center and to follow up them. Results: There was a positive correlation between different pathological grades and creatinine levels (rs = 0.275, P = 0.38), the higher the pathological grade, the higher the incidence of renal insufficiency (χ2 = 8.44, P = 0.015); the more severe the tubulointerstitial damage, the higher the incidence of renal insufficiency (χ2 = 14.15, P = 0.007); different degrees of tubulointerstitial damage and glomerulosclerosis affected the urinary protein quantification, and the difference was statistically significant (P < 0.05). The treatment effect of older is worse (P=0.030，OR=1.040，95%CI=1.004，1.078) and the effect of selecting an effective treatment is better (P = 0.002, OR = 0.618, 95% CI = 0.456, 0.839). Hypotension, cardiac involvement, decreased serum albumin, and increased TnT were independent risk factors for survival; while receiving effective treatment modalities was an independent protective factor.Conclusion: The pathological grade of renal amyloidosis is associated with the incidence of renal insufficiency, in which the degree of tubulointerstitial chronic lesions significantly affects renal function.

Studies have shown that its clinical manifestations may be related to the location and mode of amyloid deposition, glomerular deposition is mainly manifested as severe nephrotic syndrome with volume overload, while patients with tubulointerstitial or vascular deposition usually show lower proteinuria and progressive renal damage (7). Amyloid is deposited in glomeruli, renal interstitium, renal vessels, and tubules in different ways. Pathologists such as Watanabe and Saniter(8), Dikman(9), and Shiiki (10) continuously de ned and updated the deposition pattern of amyloid, and Verine et al de ned the distribution of glomerular amyloid deposits as six categories: nodular, diffuse intimal, capillary or mixed, mainly extramembranous or perimembranous, and pulmonary (11). However, a clear relationship between the degree of amyloid deposition and the severity of the clinical presentation remains uncon rmed, renal amyloid biopsy reports are not standardized, and amyloidosis progresses rapidly without a speci c treatment regimen and has a poor prognosis. Based on the above reasons, Sait et al proposed a histopathological classi cation, scoring, and grading system for renal amyloidosis to provide ideas for pathological grading as well as prognosis (12).

Diagnostic and Scoring Criteria
(1) Diagnostic criteria for renal amyloidosis: Congo red staining of renal tissue was positive and yellowgreen birefringence was observed under a polarized microscope. Electron microscopy showed a distribution of ne brillar structures 8-12 nm in diameter, unbranched, and disorganized. The diagnostic criteria for other organ involvement are shown in Table 1. (2) Histopathological classi cation and scoring criteria of renal histopathological changes: The scores included 7 aspects-Class of glomerular amyloid deposition (GAP), Percentage of glomerular amyloid deposition (GA%), renal vascular amyloid deposition (VA), renal interstitial amyloid deposition (IA), tubulointerstitial chronic lesions (I b), renal interstitial in ammatory cell in ltration (Iinf) and degree of glomerular sclerosis (GS). The degree of amyloid deposition in different tissues of the kidney was semiquantitatively scored based on the percentage of Congo red-positive material to the area of the corresponding site. The histopathological classi cation of renal amyloidosis is shown in Table 2,and speci c scoring criteria are shown in Table 3.  5. Statistical analysis SPSS 23.0 statistical software was used for data analysis. The enumeration data were expressed as n and percentage (%), the measurement data of normal distribution were expressed as x ± s, and the measurement data of skewed distribution were expressed as M (P25, P75). The means between the two groups were compared using the t-test, and the means of multiple groups were compared using one-way analysis of variance; the non-normal distribution data were analyzed using the

Renal pathological changes
Glomerular damage was predominant in 181 biopsy specimens, and three showed only interstitial or perirenal tissue damage, speci c renal pathological ndings are shown in Table 4. We found that light chain κ/λ staining was performed in 141 patients, of which 51.06% (72/141) were λ positive and 26.95% (38/141) were κ positive, with a ratio of approximately 1.89:1, and 35 were both κ and λ positive. Twenty-six patients underwent immunohistochemical staining, and only four patients underwent kappa and Lambda staining, which showed varying degrees of positivity; nine patients considering membranous nephropathy had negative PLA2R staining; and 11 patients with hepatitis B virus infection had negative HbsAg and HBeAg staining.

Correlation analysis between renal pathological changes and clinical features
Fifty-seven patients with amyloidosis nephropathy who underwent semi-quantitative scoring and grading in our center were selected and their pathological changes were correlated with urinary protein quanti cation, renal impairment and prognosis.

Correlation between renal pathological changes and renal function impairment
The rates of renal insu ciency in patients with RAPS grade I, II, III respectively were 25%, 18.75%, 58.8%, and the difference was statistically signi cant (χ2 = 8.44, P = 0.015); Spearman analysis showed that different pathological grades were positively correlated with creatinine levels (rs = 0.275, P = 0.38). Further analysis showed that the rates of renal insu ciency in patients with I b = 0, 1, 2, 3, 4 respectively were 12.5%, 0%, 34.6%, 50%, 100%, with statistically signi cant difference (χ2 = 8.44, P = 0.015). I b = 4 was signi cantly higher than other groups (P < 0.05). However, there was no signi cant difference in the incidence of renal insu ciency among patients with different scores in each group of GAP, GA%, VA, IA, GS, and Iinf.

Correlation between renal pathological changes and prognosis
The mortality rates of patients with grade I, II, III were 66.7%, 27.3%, 36.4%. The difference was not statistically signi cant. And correlation analysis showed that RAPS was not associated with mortality.
(  The clinical characteristics and prognosis were compared between the renal involvement and no-renal involvement groups. And the results showed that the length of hospital stay, serum albumin, serum calcium, glomerular ltration rate, urinary protein and creatinine were higher in the renal involvement group, and the differences had statistical signi cance (P < 0.05). While the differences in other indicators had no statistical signi cance. Therefore, patients with renal amyloidosis have a higher incidence of renal impairment than those with other organ involvement. (Table 6)

Analysis of clinical characteristics of amyloidosis patients in different years
In our study, 98 patients diagnosed in the past 5 years and 77 patients in the rst 15 years. Compared clinical characteristics and prognosis of patients in the two groups, the results showed that the age of diagnosis of amyloidosis patients in recent 5 years was higher than before, and the glomerular ltration rate was lower. The differences were statistically signi cant (P < 0.05). However, there was no signi cant difference in other laboratory parameters between two groups. But we can found that the mortality rate of this disease decreased and the cure rate increased in the past 5 years. (Table 7) 4.3 Analysis of clinical characteristics in amyloidosis patients with different prognosis The comparison of different prognostic groups showed that the serum albumin in the death group was signi cantly lower than that in the non-death group, and the BNP was higher in death group (P < 0.05). The difference in other indicators was not statistically signi cant. Therefore, decreased serum albumin and increased BNP may increase mortality. (Table 8)  coe cient results showed that hypotension, cardiac involvement, and serum albumin and cTnT at the time of diagnosis were independent risk factors for mortality (P < 0.05), and receiving treatment was an independent protective factor for mortality (P < 0.05). (Table 9)

Discussion
As the number of con rmed cases of renal amyloidosis gradually increases, the disease is no longer rare and unfamiliar. Renal amyloidosis is mainly due to deposition of amyloid in the kidney, non-speci cally manifested as edema, proteinuria and renal impairment. In our study, the number of amyloid patients diagnosed after 2014 was signi cantly higher than before, especially in patients with renal amyloidosis, with up to 89.13% of patients diagnosed in the past 5 years. The reasons are mainly related to the improvement of test methods and the continuous improvement of the understanding of the disease. After 2014, most hospitals in China carried out free light chain detection, combined with immuno xation electrophoresis and bone marrow aspiration results, for multi-level differential diagnosis, signi cantly improving the diagnosis rate. At the same time, the addition of light chain κ and λ staining items in immuno uorescence staining of pathological biopsy is also conducive to improve the diagnosis rate.  (20), while the incidence in Italy was 2.5% (21). Compared with foreign countries, relevant studies in China reported that amyloidosis nephropathy accounted for 0.89% ~ 1.27% of renal biopsy cases in the same period (22); in this study, renal amyloidosis patients accounted for 1.30% of renal biopsy cases in the same period in Shandong Province, which was lower than the incidence in foreign countries and consistent with domestic studies. Some patients did not undergo renal biopsy due to personal reasons, some patients visited other departments with extrarenal manifestations and didn't con rm by renal biopsy because of the insu cient communication between nephrology department and non-nephrology room doctors. What's more, amyloidosis nephropathy was rare before 2010, only few case reports in China. There were lack of large-sample studies about amyloidosis and deep recognition of this disease, resulting in a high rate of clinical misdiagnosis.
With renal involvement in our study, amyloid was mainly deposited in the glomeruli, mostly in the interstitial small vessel wall, and only a small part was deposited in the renal tubules and renal interstitium; there were varying degrees of tubular atrophy and interstitial brosis, with in ammatory cell in ltration, which was consistent with most studies. Six patients showed a weak positive Congo red staining, possibly due to less amyloid deposition in the early stages of the disease, or inadequate staining due to inappropriate section thickness. Eight cases which showed minimal glomerular lesions were nally diagnosed with early amyloidosis nephropathy combined with the results of special staining. Hetzel and Sayed also showed that patients with amyloidosis nephropathy may present with minimal glomerular lesions in the early stage (23,24), and the nal diagnosis of amyloidosis is con rmed after electron microscopy, multiple stains, or repeated renal biopsy. The possible reasons include: patient has both of these diseases; the morphological variations of amyloid that are indistinguishable from minor glomerular lesions on light microscopy; or amyloid deposition is too small to detect in early stages. Therefore, the initial diagnosis of amyloidosis nephropathy needs to be combined with light microscopy, special staining, electron microscopy and clinical manifestations. Moreover, the time interval between the diagnosis of minimal glomerulopathy and renal amyloidosis was up to 241 days (24). Therefore, in clinical practice, for patients with pathological results of minimal glomerulopathy, accompanied by plasmacytosis, ineffective hormone therapy, or other organ involvement, and abnormal relevant clinical indicators (light chain, immuno xation electrophoresis, etc.), it is necessary to attach great importance to whether it is amyloidosis disease and timely make a correct diagnosis.
In this study, the light microscopic results of one patient who underwent renal biopsy at an outside hospital showed membranoproliferative glomerulonephritis, but electron microscopy revealed patchy irregular lamentous brous material deposited in the renal tissue, suggesting renal amyloidosis. The other case showed glomerular mesangial ganglion hyperplasia lesions, poor therapeutic effect after several months of hormone combined with immunosuppressive agents, and lower extremity venous thrombosis, and pathological sections were consulted in the pathology department of our hospital, and Congo red staining was performed again to show positivity, con rming the diagnosis of renal amyloidosis. Therefore, not only minimal glomerular lesions, but also other pathological types can present the same situation. So electron microscopy, Congo red counterstaining, and secondary renal biopsy are all important ways to reduce the misdiagnosis rate.
Amyloid nephropathy is easily confused with other renal diseases like as broid glomerulonephritis(25), immune vibrissae glomerulopathy (26), collagen type III glomerular disease (27), and light chain deposition disease (28), all of which present with proteinuria and edema, and there is broid or granular material deposition under electron microscopy, which has been poorly recognized by doctors in the early years, leading to easy misdiagnosis in clinical practice. These diseases can be differentiated by Congo red staining and electron microscopy. Recent studies have shown the presence of DNAJB9 protein in the glomeruli of patients with broid glomerulonephritis but not in patients with several other diseases, and immunohistochemical detection of glomerular DNAJB9 has 98% sensitivity and > 99% speci city (29,30).In our study, one patient was not the age of high incidence of amyloidosis because of his young age, and there were pathological reports of broid glomerulonephritis with Congo red positivity (31), and to rule out the disease, DNAJB9 immunohistochemical detection was performed with negative results, combined with electron microscopic results to nally diagnose amyloidosis nephropathy. Therefore, the diagnosis rate can be improved by Congo red counterstaining, special immuno uorescence staining, and high magni cation observation under electron microscope.
Sait et al classi ed and scored 305 cases of renal amyloidosis, 90% of which were AA type amyloidosis, and found that pathological grade was correlated with glomerular histopathological classi cation, and the severity of glomerular amyloid deposition was positively correlated with interstitial brosis and in ammation (12). Sasatomi's study showed that serum creatinine level and pathological grade at renal biopsy were associated with accelerated progression of the disease, of which the most important prognostic factors were glomerular, tubulointerstitial and vascular lesions due to amyloid deposition(32). Yao Ying et al summarized the pathological changes of 199 patients with AL renal amyloidosis by semiintegral quantitative method and found that the degree of glomerular amyloidosis was positively correlated with the amount of urinary protein, and the incidence of renal insu ciency was higher in patients with GA and severe renal interstitial damage (22). This study found that different pathological grades were positively correlated with creatinine levels, and the higher the pathological grade, the higher the incidence of renal insu ciency; the more severe the tubulointerstitial damage, the incidence of renal insu ciency, and the relatively higher the urinary protein quanti cation; at the same time, the degree of glomerulosclerosis would also affect the urinary protein quanti cation. However, there were no signi cant correlations between the degree of amyloidosis in different renal tissues and proteinuria quanti cation or the incidence of renal insu ciency. There was also no correlation between pathological grade and mortality in patients with different amyloid nephropathy. Since the number of patients with complete renal pathology, clinical and follow-up data in our center is small, and the sample size is smaller in patients with severe pathological lesions, the relationship between the degree of amyloid deposition in different renal tissues and clinical indicators still needs to be further veri ed by large-sample study; the actual prognosis and clinical applicability of RAPS and renal amyloidosis patients need to be further validated.
In this study, the speci c classi cation was not performed, but it is worth noting that the addition of light chain κ and λ staining in the immuno uorescence staining program after 2014. In this study, 141 patients underwent light chain κ/λ staining, of which 51.06% were λ positive and 26.95% were κ positive. For one of κ or λ staining, it showed positive ( + + or + + +), while the other showed negative, then the pathological report considered AL amyloidosis; for both κ and λ, it was considered very likely to be AL; if both κ and λ were negative or weakly positive (+/-or +), further immunohistochemical staining was required to determine their classi cation. Therefore, according to the results of light chain κ/λ immuno uorescence staining, in the absence of immunohistochemistry and genetic testing, whether the clinical rough classi cation can be made in combination with the results of free light chain and immuno xation electrophoresis will provide our clinical and pathologists with great preliminary diagnosis and treatment direction.
Correlation analysis of risk factors in our study showed that blood pressure, serum albumin, eGFR, cTnT and BNP, cardiac and multiple organ involvement, and treatment methods all affected mortality. And hypotension, cardiac involvement, decreased albumin, and increased TnT were independent risk factors for survival, while receiving effective treatment was an independent protective factor for survival. For renal amyloidosis, domestic studies have shown that massive proteinuria and creatinine level at the time of diagnosis will affect the prognosis of patients (15,32). Mayo Group study indicated that creatinine level did not affect the survival time of patients, but the decrease of urinary protein quanti cation greater than 95% within 1 year may prolong the survival time of patients (33). However, in this study, it was found that decreased serum albumin was an independent risk factor for patient survival, which was related to massive proteinuria and renal impairment. So, it is crucial for doctor to pay close attention to the level of albumin, proteinuria, creatinine and eGFR.
The key to the treatment of this disease lies in stopping or reducing the production of amyloid sources and promoting clearance. At present, the removal of free light chains and small molecules mainly uses high cut-off dialysis, but inhibition of amyloid production is determined by its typing. Localized amyloidosis had a good effect of surgical treatment and a good prognosis in our study. The Mayo team followed up 413 patients with localized amyloidosis and showed that 61% of patients chose rst-line surgery, with a 10-year survival rate as high as 78% and no systemic progression (34). Patients with multiple organ involvement are usually complicated to treat and have a signi cantly higher mortality rate than single organ involvement. Chemotherapy is the main treatment for AL amyloidosis. Studies have shown that the optimal regimen is high-dose melphalan combined with autologous stem cell transplantation (ASCT), which can achieve hematological remission in about 40% of cases and control the treatment-related mortality at 10% (35). However, this surgery requires strict requirements for patients and professional centers. In our center, only 6 patients underwent ASCT, with a mortality rate of 16.67%. The mortality rate is higher than other studies possibly because of small sample. Patients not eligible for ASCT receive standard-dose chemotherapy based on alkylating agents, steroids, proteasome inhibitors, and immunomodulatory drugs(3). The most commonly regimen is bortezomib combined with dexamethasone, and studies have shown that 65% of patients on this regimen can achieve complete or partial remission, with a 5-year survival rate of 80% in intermediate-risk patients (36). 18 patients chose this regimen in our study, with the 1-year survival rate was 70.6% which was lower than that in foreign studies. Recent studies have shown that monoclonal antibodies can not only provide treatment against malignant plasma cells, but also target and remove amyloid from organs, with the advantages of high selectivity and low toxicity, becoming one of the expected clinical treatments in the future (37,38).AA amyloidosis is mainly associated with long-term in ammatory conditions (39,40), and treatment is based on the control of in ammation, clinically mainly using factors, antibiotics, colchicine and other treatments (41,42). For patients with poor e cacy or severe organ failure, organ transplantation was performed. Patients with hereditary amyloidosis are treated by liver transplantation or gene knockout.
Although treatment modalities continue to be updated, organ failure is irreversible for patients with advanced disease. Therefore, early diagnosis, classi cation and treatment are particularly important, and clinicians should continuously improve their understanding and attention to the disease.
This study is a retrospective analysis. The time span of admission for the included patients is large, the pathological and clinical data and data of some patients are missing, the conditions for medical examination in the early years are limited, and only a small part of clinical indicators can be determined. The majority of patients diagnosed in our unit are not clearly classi ed, so the number of patients with AL, AA or other types of amyloidosis cannot be determined, with the pathological, clinical characteristics and prognosis analysis among different types cannot be performed. During the follow-up, some patients are lost to follow-up due to the change of contact information. For these reasons, the survival time of some clinical indicators cannot be obtained, which has impact on the later analysis and prediction, with some limitations. However, with the continuous re nement of diagnosis and treatment techniques, as well as the increase of clinician communication and cooperation in various departments, early and correct diagnosis and effective treatment will effectively improve the prognosis of amyloidosis patients.

Conclusion
The pathological grade of renal amyloidosis is associated with the incidence of renal insu ciency, in which the degree of tubulointerstitial chronic lesions signi cantly affects renal function. The relationship between the degree of starch deposition in different renal tissues and the levels of proteinuria and renal function still needs to be explored in large sample studies. At the same time, clinicians should pay more attention to the relevant risk factors to improve the early diagnosis rate and improve the prognosis of patients. Declarations