The utility of repeat kidney biopsy in systemic immunoglobulin light chain amyloidosis

Abstract

Background

The diagnostic utility of repeat kidney biopsy in AL amyloidosis patients in complete (CR) or very good partial hematologic response (VGPR) but with renal organ relapse is not clear.

Methods

We present 8 patients with AL amyloidosis who had a repeat kidney biopsy performed.

Results

AL amyloidosis was initially diagnosed by a kidney biopsy. All patients had a favorable response to treatment (CR/VGPR) and 5 of them also had initially a renal organ response. A repeat kidney biopsy was done due to gradual deterioration of kidney function and/or proteinuria while maintaining a hematologic response. Repeat kidney biopsies showed findings consistent with amyloid deposits in all patients. Seven patients had renal progression with 4 of them requiring dialysis initiation. Only one patient had a favorable renal outcome. This patient had subacute kidney injury with decreasing proteinuria and was found to have granulomatous interstitial nephritis in addition to amyloid deposits and responded well to steroid treatment with rapid improvement in renal function.

Conclusions

In AL amyloidosis patients who achieve a favorable hematologic response to treatment (CR/VGPR) but subsequently develop worsening renal insufficiency or proteinuria, a repeat kidney biopsy should generally not be performed. Amyloid deposits persist in the kidneys even after successful hematologic treatment and it is impossible to differentiate between new versus old amyloid deposits, which makes performing a repeat kidney biopsy unnecessary in most cases. Demonstration of amyloid deposits on repeat kidney biopsy would not aid in the decision making regarding re-initiation of hematologic treatment. A kidney biopsy should be considered only in cases when a specific alternative diagnosis is suspected.

Introduction

Renal involvement in systemic immunoglobulin light chain (AL) amyloidosis is found in over two-thirds of patients [1–3] and usually manifests as 24-hour urine protein excretion of more than 0.5 gram and/or renal insufficiency.

Treatment for AL amyloidosis has significantly improved in the past two decades leading to longer patient survival. Successful therapy is characterized by hematologic response (complete response (CR) or very good partial response (VGPR)) followed by organ response [4]. Hematologic response and organ response criteria have been unified by international investigators. Hematologic CR is defined as negative serum and urine immunofixation and κ:λ free light chain ratio within reference range. Hematologic VGPR is defined as the difference between serum level of involved and uninvolved free light chains (dFLC) of less than 40mg/dL. Partial response (PR) is defined as dFLC of >50% [5]. Considering the fact that free light chains (FLCs) are cleared by the kidneys, impaired kidney function can affect their serum levels and makes it difficult to differentiate between VGPR and PR in patients who have abnormal renal function. Renal organ response is defined as stabilization or improvement of renal function with more than 30% decrease in proteinuria and has been shown to correlate with lower annual risk of end stage renal disease (ESRD) [6–9].

Current therapies in the treatment of AL amyloidosis focus on targeting plasma cells and not the amyloid deposits themselves in the affected organs. Nonetheless, achievement of hematologic response usually leads to improvement in markers of organ function, albeit in a delayed manner. After achieving a hematologic response the median time to organ response is around 10 months [10].

Despite high rates of hematologic and renal organ response to first line therapy, the disease often recurs [11,12], potentially leading to worsening renal function as a result of continued amyloid deposition. Increasing proteinuria and/or worsening renal function after successful treatment always raise the possibility of recurrent amyloidosis but hematologic relapse is not always obvious. A subset of the patients has renal progression with normal monoclonal FLC levels or the levels are elevated but in the setting of impaired renal function. In these cases, it is not clear if there is a small amount of circulating amyloidogenic light chain below the detection level or if the worsening of renal disease represents progression of the original, irreversible renal organ damage. Recently, newer methods for evaluation of hematologic response such as flow cytometry and next generation gene sequencing used primarily in multiple myeloma (MM) have enabled detection of malignant clones that are undetectable by standard methods [13,14]. These clones are defined as minimal residual disease (MRD), and their presence in patients with MM was associated with inferior disease-free and overall survival [13,14]. When used in AL amyloidosis patients, next generation flow cytometry revealed that 45–60% of those previously classified as sustained CR are in fact MRD positive [15,16].

The utility of a repeat kidney biopsy in AL amyloidosis patients after successful hematologic treatment with worsening renal function and/or proteinuria is not clear. In these patients the most likely diagnosis is renal amyloidosis although other possible etiologies cannot be ruled out with absolute certainty. Differential diagnosis frequently includes diabetic nephropathy, secondary focal segmental glomerulosclerosis, a de-novo glomerular disease or interstitial nephritis. In these cases, a definitive diagnosis at times requires a kidney biopsy.

In the literature there are a few reports describing repeat kidney biopsies that have been performed in patients with AL amyloidosis for uncertain indications, such as follow up of response to treatment [17–21] or in order to rule out the presence of amyloid fibrils in the setting of re-appearance of proteinuria or elevation in serum creatinine [22]. Here we present 8 patients with AL amyloidosis in whom a repeat kidney biopsy was performed for various reasons in order to evaluate the diagnostic utility of this test.

Materials and methods

We retrospectively analyzed the clinical outcomes and renal biopsy findings of 8 patients with AL amyloidosis who were followed at the Amyloidosis Center at Boston University School of Medicine between 2001–2018 and had a repeat kidney biopsy performed during the course of their disease to delineate the etiology of worsening renal function and/or proteinuria. The study was approved by the Institutional Review Board in accordance with the Declaration of Helsinki and the Health Insurance Portability and Accountability Act guidelines. Informed consent has been obtained from all patients for use of their clinical and laboratory information as specified in the International Committee of Medical Journal Editors Recommendations (ICMJE). Glomerular filtration rate (GFR) was estimated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation [23]. Histological findings were noted from biopsy reports.

Results

All patients were evaluated at the Amyloidosis center and diagnosed with AL amyloidosis by a first kidney biopsy performed between the years 2001–2016. Six patients were females and 2 were males, with a median age at diagnosis of 57.4 years (range, 37.1–71.5). Median estimated GFR (eGFR) at diagnosis was 62 mL/min/1.73m² (range, 21–107) and median 24-hour urine protein excretion was 7.8 g (range, 3.3–15.5) (Table 1). Seven patients had an associated λ clonal plasma cell dyscrasia and one patient had κ light chain predominance. The latter patient was diagnosed with Waldenstrom macroglobulinemia (WM) 2 years prior to the diagnosis of AL amyloidosis. Six patients received high dose melphalan/stem cell transplantation (HDM/SCT) as initial treatment for AL amyloidosis, one received bortezomib-dexamethasone and one was treated initially with rituximab and subsequently with oprozomib and rituximab-bendamustine for WM (Table 2). Four patients achieved a hematologic CR and 4 achieved VGPR to initial therapy. Five patients achieved renal organ response to first line therapy.

Table 1.

Clinical features at diagnosis and tie of repeat kidney biopsy

Patient #	Gender	Dx	LC clonality	Age at Dx (years)	Time of 1st Bx from Dx (years)	Cr at time of 1st Bx (mg/dL)	eGFR at time of lst Bx (mL/min/1.73m2)	Proteinuria at time of 1st Bx (gram)	Free K:λ ratio at time of 1st Bx (normal range: 0.26– 1.65)	Cr at time of 2nd Bx (mg/dL)	eGFR at time of 2nd Bx (mL/min/1.73m2)	Proteinuria at time of 2nd Bx (gram)	Free K: λ ratio at time of 2nd Bx (normal range: 0.26 –1.65)
1	F	AL amyloidosis	λ	53.5	0	1	64	7.8	0.29	3.1	16	1.3	1.5
2	F	AL amyloidosis	λ	37.1	0	1	72	15.5	0.25	2.4	23	4.7	0.98
3	F	AL amyloidosis	λ	41.7	0	0.7	107	5.3	0.06	0.9	75	13	0.29
4	F	AL amyloidosis	λ	48.0	0	1.7	35	5.9	2.9	2.7	20	7	0.61
5	F	AL amyloidosis	λ	71.5	0	0.8	74	3.3	0.27	2.5	17	2	0.98
6	M	WM with AL amyloidosis	κ	68.8	2.0	2.9	21	20	4.6	4.1	14	11	3.28
7	F	AL amyloidosis	λ	66.3	0	1.7	31	13	0.73	2.5	19	1.5	0.93
8	M	AL amyloidosis	λ	61.4	0	1.3	59	9.7	0.17	1.2	63	18	0.52

Bx, biopsy; Cr, creatinine; Dx, diagnosis; eGFR, estimated glomerular filtration rate; LC, tight chain

Table 2.

Clinical features at the time of repeat kidney biopsy

Patient #	Type of treatment	Response to hematologic treatment	Renal response	Time from 1 st Bx to 2nd Bx (years)	Reason for 2nd kidney Bx	BMB at time of 2nd Bx
1	HDM/SCT	CR	YES	3.0	AKI with Cr 1.3-->4 thought to be AIN from furosemide	No evidence of PCD, 10– 15% plasma cells with no clear predominance, Congo red negative
2	HDM/SCT	CR	YES	9.3	Increasing Cr and proteinuria	No evidence of PCD, 5–10% plasma cells with no predominance
3	HDM/SCT	CR	YES	7.7	Increasing proteinuria	No evidence of PCD, <5% plasma cells with no predominance
4	HDM/SCT	VGPR	YES	3.1	Increasing Cr	Congo Red stain in blood vessels and focally in interstitium, slight predominance for λ
5	HDM/SCT	VGPR	YES	11.3	Increasing Cr, slight increase proteinuria	Not done
6	Rituximab, orpozomib, bendamusitine	VGPR	NO	3.5	Increasing Cr and proteinuria	Congo red positive in blood vessels, 5% plasma cells with slight K predominance
7	Bor-dex	CR	NO	2.4	Increasing Cr	No evidence of PCD, 0.7% polytypic plasma cells
8	Len-Dex, HDM/SCT, len- Dex, Bor-Dex	VGPR	NO	5.2	Increasing proteinuria	Congo red positive in blood vessels, 5–10% plasma cells with slight λ predominance

AIN, acute interstitial nephritis; AKI, acute kidney injury; BMB, bone marrow biopsy; Bor-dex, bortezomib-dexamethasone; Bx, biopsy; Cr, creatinine; CR, complete response; HDM/SCT, high dose melphalan/stem cell transplantation; Len-dex, lenalidomide-dexamethasone; PCD, plasma cell dycrasia; VGPR, very good partial response

In all cases the first kidney biopsy was performed at the time of diagnosis. Biopsy findings are summarized in Table 3. All kidney specimens (both first and second biopsies) demonstrated Congo red positivity on light microscopy (LM) with green birefringence under polarized light. Patient #3 was initially diagnosed with amyloidosis based on EM findings showing fibrils consistent with amyloidosis. Repeat Congo red staining was subsequently performed on LM specimens and was found to be positive. Typical amyloid fibrils were seen on all EM specimens except on the first specimen obtained from patient #8 in which amyloid deposition was reported to be mainly in the vascular compartment. The repeat kidney biopsy was done after a median time of 4.3 years (range, 2.4–11.3) from the first biopsy, mostly due to gradual deterioration in kidney function and/or increase in 24-hour proteinuria, in the setting of normal FLC ratio (Table 2). Median eGFR at the time of repeat kidney biopsy was 20 mL/min/1.73m² (range, 14–75) (Table 1). Biopsy samples showed varying degree of interstitial fibrosis and tubular atrophy (IFTA) (Table 3). One patient presented with acute kidney injury (AKI) and pyuria with the pre-biopsy diagnosis of allergic interstitial nephritis (AIN) due to furosemide. At that point, 2 years after HDM/SCT, proteinuria had decreased from 7.8 g to 1.2 g and she was categorized as being in CR with renal organ response. Subsequently she was found to have granulomatous interstitial nephritis in addition to λ amyloid deposition in the glomeruli and interstitium, and was diagnosed with sarcoidosis.

Table 3.

Comparison of first and second kidney biopsy findings

Patient #	Biopsy category	1st biopsy	2nd biopsy	Outcomes
1	Amyloid distribution and extent(Congo red)	Glomeruli- moderate Interstitium- mild Vascular- mild	Glomeruli - moderate-severe Interstitium - mild Vascular- mild	Eventually diagnosed with sarcoidosis. Cr decreased to 1.3 with steroid treatment.No further hematologic treatment was given.
	IFTA	Mild	Mild-moderate
	Glomerulosclerosis	None	None
	Arteriolosclerosis	Mild	Mild-moderate
	Other lesions	None	Granulomatous interstitial nephritis
	IF	λ+ in glomeruli and arteriols, focal C3+ in TBM	λ + in glomeruli and arteriols, focal C3(+2) in TBM and arteriols
	EM	Non-branching fibrils in mesangium	Non-branching fibrils in mesangium
2	Amyloid distribution and extent(Congo red)	Glomeruli- severe Interstitium- mild Vascular- mild	Glomeruli - severe Interstitium - severe Vascular- severe	Kidney biopsy complicated by severe bleeding, hypotension and need for acute dialysis with renal recovery back to baseline. Progressed to ESRD within 13 months of repeat kidney biopsy. No further hematologic treatment was given.
	IFTA	Mild	Moderate
	Glomerulosclerosis	None	Moderate
	Arteriolosclerosis	Mild	Mild
	Other lesions	None	None
	IF	Diffuse λ staining in mesangium and focally in interstitium	λ staining (+2) in glomeruli and interstitium, IgG - trace
	EM	Non-branching fibrils (9–14 nm) in mesangium and TBM	Extensive mesangial expansion by small non-branching fibrils. Severe epithelial foot process efacement.
3	Amyloid distribution and extent(Congo red)	Glomeruli- mild Interstitium - none Vascular- none	Glomeruli - moderate Interstitium - mild Vascular- mild	Received bortezomib/ dexamethasone and revlimid/ dexamethasone post 2nd kidney biopsy. Progressed to ESRD within 6 months from repeat kidney biopsy.
	IFTA	None	Mild
	Glomerulosclerosis	Mild	Mild
	Arteriolosclerosis	Mild	Mild
	Other lesions	None	Many foamy macrophages in interstitium
	IF	Predominanti positive λ staining in mesangium, + amyloid P in mesnagium, (+2) C3 mesangial granular staining	(+3) λ staining in mesangium, (+2) IgG, C3, (+3) amyloid P
	EM	Mild mesangial expansion with 12nm randomly oriented fibrils	Mesangial expansion with 8–10nm randomly oriented fibrils
4	Amyloid distribution and extent(Congo red)	Glomeruli - mild Interstitium- none Vascular- severe	Glomeruli - mild-moderate Interstitium - severe Vascular - severe	P rogressed to ESRD within 6 months from repeat kidney biopsy. Also had decreasing cardiac function with no change in hematologic markers. No further hematologic treatment was given.
	IFTA	Mild-moderate	Severe
	Glomerulosclerosis	Mild	Moderate
	Arteriolosclerosis	None	Mild
	Other lesions	Lymphocytic interstitial infiltrate	Lymphoplasmacytic interstitial infiltrate
	IF	(+3) λ mesangial and peri-vascular staining	(+3) λ mesangial, interstitial and vascular staining , (+1–2) C3 segmental glomerular staining
	EM	Segmental mesangial expansion overlapping non-branching fibrils	Mesangial expansion overlapping non-branching fibrils, mild epithelial foot process effacement
Patient #	Biopsy category	1st biopsy	2nd biopsy	Outcomes
5	Amyloid distribution and extent (Congo red)	Glomeruli - mild Interstitium - mild Vascular - mild	Glomeruli - severe I n terstitium- severe Vascular- severe	Cr 2.5, eGFR 17ml/min/1.73m2 at last follow up. No further hematologic treatment was given.
	IFTA	Mild	Severe
	Glomerulosclerosis	Moderate	Moderate-severe
	Arteriolosclerosis	Moderate	Moderate-severe
	Other lesions	None	None
	IF	Nonspecific mesangial (+1) IgG and IgM, no light chain predominance	(+2) mesangial, tubulointerstitial and vascular staning for λ
	EM	Fibrillary deposits present in mesangial matrix, interstitial blood vessels and interstitium	Fibrillary deposits present in mesangial matrix, interstitial blood vessels and interstitium, mild segmental foot process effacement
6	Amyloid distribution and extent (Congo red)	Glomeruli -moderate Interstitium - none Vascular - mild	Glomeruli - moderate Interstitium - mild Vascular - moderate	Cr 4.1, eGFR 13ml/min/1.73m2 at last follow up. Recommended further treatment for lymphoplasmacytic lymphoma with bortezomib, dexamethasone and rituximab.
	IFTA	Moderate	Moderate -severe
	Glomerulosclerosis	Mild	Moderate
	Arteriolosclerosis	Severe	Moderate-severe
	Other lesions	Acute tubular injury with focal epithelial necrosis	Lymphocytic interstitial infiltarte
	IF	Masnagial IgM(+3), C3 (trace) and K (+3). Granular fine C3 (+1) in TBM	No glomeruli - performed on pronase digested tissue. No light chain predominance
	EM	Extensive mesangial deposition of non-branching randomly arranged fibrils (8nm), focal epihtelial foot process effacement	Mesangial deposition of non- branching randomly arranged fibrils (8–12nm), extensive epihtelial foot process effacement
7	Amyloid distribution and extent (Congo red)	Glomeruli- moderate Interstitium - mild Vascular - mild	Glomeruli - modearte Interstitium - moderate Vascular - none	Cr 2.5, eGFR 19ml/min/1.73m2 at last follow up. No further hematologic treatment was given.
	IFTA	Mild	Moderate
	Glomerulosclerosis	None	None
	Arteriolosclerosis	Mild	Mild-moderate
	Other lesions	None	None
	IF	(+4) λ staining on mesangium and focally on vessel walls	(+4) λ staining on mesangium and focally in interstitum
	EM	No glomeruli present, 10nm randomly arranged fibrils in the interstitum and vessels	10nm randomly arranged fibrils in mesangium and interstitium
8	Amyloid distribution and extent (Congo red)	Glomeruli - mild Interstitium - none Vascular - moderate-severe	Glomeruli - moderate Interstitium - moderate Vascular - moderate	Progressed to ESRD within 10 months from repeat kidney biopsy in spite of receiving several lines of hematologic treatment. Had comlications due to myelosuppression.
	IFTA	Mild	Mild
	Glomerulosclerosis	Mild	Mild
	Arteriolosclerosis	Modrate-severe	Moderate
	Other lesions	None	None
	IF	No significant glomerular staining. (+1) granular IgG and trace λ	(+3) λ mesangial, interstitial and vascular staining , (+1) IgG, IgM and K in vessels
	EM	No fibrils seen in the limited sample	Randomly oriented, non-branching fibrils (10nm) involving mesangium, GBM and interstitium

Cr, creatinine; eGFR, estimated glomerular filtration rate; ESRD, end stage renal disease; FLC, free light chain; GBM, glomerular basement membrane; IF, immunofluorescence; IFTA, interstitial fibrosis and tubular atrophy; EM, electron microscopy; PCD, plasma cell dyscrasia; TBM, tubular basement membrane

A bone marrow biopsy (BMB) was done at the time of repeat kidney biopsy in 7 of the 8 patients. All BMB had 5–10% plasma cells, 2 with λ, 1 with κ predominance, and 4 with no predominance. Two patients had positive Congo red staining in bone marrow blood vessels.

Three of the 8 patients (#3, 6, 8) received additional hematologic treatment following the second kidney biopsy without achieving renal organ response. Four patients advanced to ESRD 6–13 months from the time of the second kidney biopsy. Three more patients progressed to advanced chronic kidney disease (CKD) with eGFR< 20mL/min/1.73m² not yet requiring dialysis initiation. The only patient who had a positive renal outcome was the patient who had AKI due to granulomatous interstitial nephritis with improvement in her creatinine after treatment for sarcoidosis (down to 1.3 mg/dL from 3 mg/dL).

One patient (#2) developed severe bleeding with hypotension post biopsy and required several blood transfusions, embolization and acute dialysis for 48 hours.

Discussion

This study presents 8 patients with AL amyloidosis and renal involvement who underwent repeat kidney biopsy at various time points during their disease course while being in CR or VGPR based on hematologic parameters. Amyloid deposits were present in both kidney biopsies in all cases.

There are only a handful of cases of repeat renal biopsies in amyloidosis patients described in the literature. Nine repeat kidney biopsies in 8 cases (one patient had three kidney biopsies) were reportedly done for the purpose of determining whether complete hematologic response to treatment, accompanied by decline in proteinuria, was associated with disappearance of amyloid deposits from the kidney [17–21]. In all of these cases, AL amyloid deposits were still present in the repeat kidney biopsy specimens. Of note, one of these reports of 2 patients from Mayo Clinic specifically described the disappearance of the immunoglobulin component from the deposits with persistence of serum amyloid P (SAP) and apoE as evaluated by IF and laser microdissection followed by mass spectrometry [19]. SAP has been shown to prevent proteolytic degradation of amyloid fibrils in vitro and been used for detection of amyloid distribution and estimation of the amount of amyloid in visceral organs throughout the body (SAP scan) [24,25]. In all our repeat biopsies there were still detectable light chain component by IF.

As technologies for detecting ever more minute levels of aberrant plasma cell clones in the bone marrow or circulating amyloidogenic light chains in the blood have become available, the use of MRD assessment in AL amyloidosis patients has grown into an emerging area of interest [13,14,26,27]. It is conceivable that despite hematologic CR by current consensus criteria a malignant clone is still present, causing continued amyloid deposition in the kidney and leading in turn to worsening renal parameters.

One other case report by Roth et al. described 2 patients with AL amyloidosis who were diagnosed by a kidney biopsy, treated with HDM/SCT and achieved a CR. A repeat kidney biopsy was performed due to an increase in creatinine and urine protein excretion and was reported in both patients as progressive renal amyloidosis (described as significant increase in the size of amyloid deposits but without morphometric analysis), despite persistence of complete hematologic response with no detectable serum and/or urine monoclonal immunoglobulins [22]. The authors attributed the progression of amyloid deposits to a type of treatment failure, which could also be attributed to MRD positivity. It has been suggested that the process of amyloid formation can be accelerated by the presence of pre-formed fibrils which capture and catalyze the conversion of monomeric precursors, even at very low concentrations, into misfolded, toxic, and aggregation-competent structures [28]. An alternative explanation would be that amyloid deposits, which are not evenly distributed in the kidney, were simply more abundant in the specimen from the repeat biopsy. Also, at this point morphometric analyses of the amyloid deposits are not done routinely and comparison of the size of the deposits has not been evaluated in a vigorous scientific fashion or correlated with clinical outcome. Regardless of the mechanism, in all the aforementioned cases, whether there was renal organ response or not, amyloid fibrils clearly did not disappear.

Van Gameren et. al studied the effect of chemotherapy on light chain content in fat tissue in repeat fat biopsies in 38 patients with AL amyloidosis (26 complete responders, 6 partial responders and 6 non-responders). At the median follow up of 1.6 years (range, 0.3–9) the concentration of amyloid in fat tissue regressed significantly in complete responders and partial responders but not in non-responders [29]. However, only about 30% of the patients with CR showed complete normalization of fat FLC concentration, which shows that even in fat tissue it takes a long time to achieve complete resolution of amyloid deposits after successful treatment. Still, based on currently available data, it is likely that amyloid deposits in fat tissue are more readily dismantled than in kidneys due to factors that are associated with specific organ tropism. Vital organ response was not correlated to the concentration of FLC in fat tissue in this study.

The 8 patients presented here had persistence of amyloid deposition in repeat kidney biopsy samples. Most of them had good kidney function at the time of diagnosis (median eGFR 62 mL/min/1.73m²,range 22–92) and they achieved a favorable hematologic response to treatment. Five of them also had renal organ response initially with stabilization of renal function and more than 30% decrease in proteinuria. However, at the time of repeat kidney biopsy, median eGFR had decreased to 20 mL/min/1.73m² with 4 of them advancing to ESRD within 6–13 months, suggesting organ disease relapse.

Of note, the percentage of IFTA in kidney biopsy specimens usually correlates with progression to ESRD [30]. In our 8 patients, the amount of IFTA (mild, moderate or severe) at the time of the second biopsy had a good correlation with progression to ESRD with a few exceptions. For example, at the time of the second biopsy, patient #5 had severe IFTA and concordantly had eGFR as low as 17mL/min/1.73m², although at the time of last follow up he had not yet reached ESRD without any additional hematologic treatment. Patients #3 and #8 on the other hand, had mild IFTA at the time of the second biopsy, and both of these patients had eGFRs >60mL/min/1.73m². Notwithstanding the foregoing and despite receiving additional hematologic treatment, both patients progressed to ESRD within 6 and 10 months, respectively. These discrepancies could be attributed to the fact that the biopsy sample has a margin of error and it might not accurately represent the true degree of IFTA in the whole kidney. It is also possible that circulating light chains (MRD) were present in these cases or amyloid deposits themselves in addition to structural damage somehow accelerate the deterioration of kidney function by a mechanism that has not yet been fully identified. It has been suggested that pre-fibrillar amyloidogenic light chains (also known as oligomers or aggregates) are inherently cytotoxic [31,32]. One must also keep in mind that eGFR is an estimation derived from population based studies, however it does not take into account individual factors such as body weight, muscle mass or renal tubular dysfunction, therefore it might under- or over-estimate true renal function.

One patient, who presented with subacute renal failure while in CR and had decreasing proteinuria, was found to have granulomatous interstitial nephritis and was subsequently diagnosed with sarcoidosis. Despite having a different etiology for her kidney function impairment, which later resolved with prednisone, amyloid deposits were also seen in the repeat kidney biopsy specimen. In this case the suspicion for a different etiology was very high given the acute nature of the renal failure, and the presence of sterile pyuria with white blood cell casts.

Importantly, although a kidney biopsy is not considered a high-risk procedure, it is not completely benign and may result in complications such as bleeding which can be severe. It may even, as a worst case scenario, lead to nephrectomy. Post-biopsy bleeding was evaluated in a retrospective study of 101 patients with systemic amyloidosis and 180 patients with no evidence of amyloidosis who served as controls. Similar bleeding risk was observed in both groups: 9.9% of patients with amyloidosis compared to 10.6% in the control group (p=0.8). Major bleeding requiring blood transfusion or intervention was also not statistically different between the groups but there was a trend towards higher risk in the amyloidosis group (4% vs. 2.1%, p=0.4) [33]. In our series of 8 patients, one repeat kidney biopsy was complicated by severe bleeding and hypotension requiring blood transfusions, local embolization and need for short term dialysis.

Clearly, amyloid deposits persist in the kidneys even after successful hematologic treatment [17–22]. At this point in time, it is impossible to differentiate between new versus old amyloid deposits, which makes performing a repeat kidney biopsy unnecessary in most cases. Demonstration of amyloid deposits on repeat kidney biopsy does not add new information and therefore would not aid in the decision making regarding re-initiation of hematologic treatment. In these cases we suggest relying on clinical parameters. In the future, it may be possible to use more sensitive detection methods to find the malignant clone in the bone marrow or circulating amyloidogenic light chains that are below the current detection level. A kidney biopsy should be considered only in cases when a specific alternative diagnosis is suspected.

Abbreviations

AIN

allergic interstitial nephritis

AKI

acute kidney injury

AL

light chain

BMB

bone marrow biopsy

CKD

chronic kidney disease

CR

complete response

EM

electron microscopy

ESRD

end stage renal disease

GFR

glomerular filtration rate

FLC

free light chain

HDM/SCT

high dose melphalan/stem cell transplantation

IFTA

interstitial fibrosis and tubular atrophy

LM

light microscopy

MM

multiple myeloma

MRD

minimal residual disease

PR

partial response

SAP

serum amyloid protein

VGPR

very good partial response

WM

Waldenstrom macroglobulinemia