Introduction

Monoclonal immunoglobulin deposition disease (MIDD) is characterized by the deposition of monoclonal light and/or heavy chains within glomerular, tubular, and vessel wall basement membranes. Three subtypes of MIDD have been described and they are subdivided according to the composition of monoclonal protein deposited. The most common form of MIDD is light chain deposition disease (LCDD), where either kappa or lambda light chains are deposited in tubular, glomerular, and vessel wall basement membranes. Much less common than LCDD are heavy chain deposition disease (HCDD) and light and heavy chain deposition disease (LHCDD), which show basement membrane accumulation of a monoclonal immunoglobulin heavy chain, either with a corresponding monoclonal light chain (LHCDD) or without an accompanying light chain (HCDD). ^, In contrast to the fibrils seen in amyloidosis, the deposits of MIDD lack substructure. In the past, the term monoclonal immunoglobulin deposition disease has been used as a generic term to encompass a wide variety of paraprotein-related diseases including light chain amyloid and cast nephropathy. In this chapter, we restrict ourselves to the currently accepted narrower usage of the term as defined earlier, or what has been historically called Randall-type monoclonal immunoglobulin deposition disease . For much of the chapter we combine our discussion of patients with LCDD, LHCDD, and HCDD, referring to them as patients with MIDD. Notable differences between subtypes of MIDD are pointed out where appropriate and summarized in Table 9.1 .

Epidemiology of monoclonal immunoglobulin deposition disease

MIDD is a relatively rare disease. The incidence of pure MIDD (MIDD in the absence of light chain cast nephropathy or amyloidosis) in native kidney biopsies has been reported to range from 0.32% in a study from Columbia University Medical Center in New York up to 0.7% of renal biopsies processed at the Mayo Clinic. Overall, MIDD appears to be approximately half as common as amyloidosis. In an autopsy series of patients with multiple myeloma (MM), LCDD was found in 5% of patients, whereas amyloid was detected in 11%. Similarly, out of the 1078 biopsies studied at Columbia University between 2000 and 2014 with paraprotein deposition in the kidney, pure MIDD was identified in 209 (19%), compared with amyloidosis in 451 (41%), and light chain proximal tubulopathy in only 54 (5%).

The association of MIDD with underlying hematologic conditions has varied across different studies as our understanding of plasma cell dyscrasias has evolved and new definitions of MM have emerged. In a recent report by Kourelis et al., 57% of patients with MIDD had either smoldering MM or MM. The literature indicates that MM is present in roughly half of patients with LCDD and LHCDD, but the percentage has been reported as high as 58% to 65% for those with LCDD. ^{, ,} In contrast, HCDD has a much lower reported incidence of MM compared with LCDD and LHCDD, with reports ranging from 19% to 33%. ^{, ,} After the introduction of the term monoclonal gammopathy of renal significance ( MGRS ), virtually all patients with MIDD have been shown to have a plasma cell dyscrasia. ^{, ,}

Patient characteristics of those diagnosed with MIDD are notable for mean age, typically in the late 50s, although ages have ranged from patients in their 20s to patients over 90 years. ^{, , ,} In the series by Nasr et al., approximately one-third of patients with MIDD were under 50 years of age. In patients with LCDD, there appears to be a male predominance, with approximately twice as many men as women. This is in contrast to HCDD, where men and women appear to be equally affected. ^{, , ,} Race is not reported in all studies; however, in those large case series where it was, patients were white in more than 80% of the cases. ^{, ,} Of interest, the only study with a significant number of black patients (22%) was from a single institution in New York, and four of five of those patients had either HCDD or LHCDD. The only large case series from Asia, reported to date by Li et al., examined 48 patients, all of whom were Chinese, and age and gender data were similar to the trends described previously. It is difficult to draw meaningful conclusions regarding any differences in racial involvement, given that most of the reports have come from geographic areas where there is not a wide range of racial diversity and the majority of patients studied were white.

Clinical manifestations of monoclonal immunoglobulin deposition disease

Whereas MIDD can present as a multisystem disease, renal manifestations typically predominate. More than 83% of the patients reported with MIDD and kidney disease had renal insufficiency based on serum creatinine greater than 1.2 mg/dL or chronic kidney disease (CKD) stage 2/3. Many patients experience a rapid decline in renal function. ^{, , ,} Proteinuria is present in almost all patients with MIDD, and about 40% to 50% of patients will have nephrotic range proteinuria. ^{, ,} The prevalence of nephrotic range proteinuria and nephrotic syndrome seems to be higher in those with HCDD. In contrast to amyloidosis, hypertension is widely present at the time of diagnosis in patients with MIDD, with two large series reporting hypertension in approximately 80% of patients. ^, However, reports of hypertension in LCDD ^, may be closer to 50%, whereas patients with HCDD have a higher incidence with nearly 100% of them being affected. ^{, , ,} Incidence of microscopic hematuria is reported to range from 60% to 90% of the cases of MIDD and also appears to be somewhat more common in patients with HCDD. ^{, ,} Hypocomplementemia was absent in all 12 patients with LCDD reported by Lin et al., but was present in three of six of their HCDD patients. Bridoux et al. reported low C3 in five of their 15 patients with HCDD; four of these patients had γ1-HCDD and the other had γ3-HCDD, correlating with the known complement fixing abilities of immunoglobulin (Ig)G subtypes 1 and 3. Of note, a recent report from China did show C3 hypocomplementemia in up to one-third of patients with LCDD.

Whereas renal manifestations in MIDD typically predominate, extrarenal involvement in multiple organ systems is well described. From the first descriptions of LCDD by Randal et al. in the 1970s, it was demonstrated that visceral deposition in multiple organs was a feature of the disease. LCDD has been described to involve the heart, liver, lungs, pancreas, thyroid, skin, lymph nodes and spleen, muscles, and the gastrointestinal and neurologic systems. As reported by Ganeval et al., when extrarenal manifestations were specifically evaluated for, they were found in all MIDD patients. Light chain deposits were found in the blood vessels walls of most organs. Heart and liver involvement are the most commonly described extrarenal sites of MIDD involvement. Heart disease is reported in up to 80% of cases with LCDD; however, whether the etiology of the cardiac disease can be attributed to MIDD is unknown, because most patients do not undergo heart biopsy or have evidence of an infiltrative process in the heart. A significant portion of the cardiac disease in MIDD patients may relate to confounding factors, specifically age and renal impairment. ^{, ,} In the past, cardiac involvement may have been historically underdiagnosed because of lack of usage of immunohistologic and electron microscopic methods in endomyocardial biopsies. Lack of uniformity in nomenclature also hampers our ability to estimate cardiac involvement because in the cardiac literature, some studies refer only to cardiac nonamyloidotic immunoglobulin deposition disease (CIDD), light-chain cardiomyopathy, or simply describe cases as cardiomyopathies with light chain deposition disease. Toor et al. reported the experience of one MM referral center where less than 2% of the patients referred for evaluation of MM underwent cardiac biopsy. Of those, 20% (six patients) had CIDD. In total, eight cases were analyzed but four of eight patients also had proven concomitant amyloidosis in other organs. Interestingly, none of the patients had any symptoms related to heart failure. The median age for these patients at presentation and diagnosis was somewhat earlier (median age of 49.5 years, range 40–64 years). In general, in patients with cardiac manifestations, the most common clinical features include congestive heart failure, usually with a restrictive pattern, arrhythmias, and conduction anomalies. In two series in which echocardiography was reported, 10/13 patients had left ventricular hypertrophy. ^,

In one of the earliest case series, liver deposits were found in almost all of the patients who underwent hepatic tissue analysis. Hepatomegaly and moderate abnormalities in liver function tests are often observed; however, more rarely, cases of portal hypertension, hepatic insufficiency, and death from fulminant liver failure have also been described. Histologic features of liver involvement include abnormal chromophilic material along sinusoids, in vascular walls, and in basement membranes of biliary ductules under light microscopy. Portal areas were enlarged by fibrosis and usually contained abnormal material. The distribution of lesions varied from patient to patient. ^, The burden of deposits in liver biopsies has been described as moderate to severe, but mild lesions with nearly normal sinusoids are also seen. ^,

Pulmonary symptoms and involvement of the respiratory system are less common. Respiratory symptoms can be heterogeneous, involving multiple sites including the large airways. Pulmonary involvement can be very indolent and patients may remain asymptomatic for several years, although severe cases requiring lung transplant have been reported. Histologically, pulmonary involvement can be characterized by either diffuse involvement or a more nodular distribution. Newer images with high resolution computed tomography scans show that almost all patients with pulmonary involvement have thin-walled cysts, usually with pulmonary nodules.

Neurologic involvement in MIDD is common, with up to 20% to 30% of patients having peripheral neuropathy caused by deposits along the nerve fibers. ^{, ,} Central nervous system involvement is rare likely because of limited crossing of paraproteins through the blood-brain barrier. However, “aggregomas”—a term coined by Rostagano et al., and described as localized tumoral masses of Congo-red-negative monoclonal nonfibrillar Ig light chain, have also been reported. ^, Isolated “aggregomas” without evidence of any plasma cell dyscrasia and presenting with large, intracerebral lesions exerting significant mass effect have been recently reported. ^,

Histopathologic features of renal involvement

The light microscopic findings in LCDD, LHCDD, and HCDD overlap considerably. Whereas earlier series describe nodular sclerosing glomerulopathy as essentially universal in all forms of pure MIDD, subsequent series have found nodular sclerosis in only 61% of cases, with the remaining cases showing only mild or no mesangial sclerosis. Notably, in HCDD, nodular mesangial sclerosis appears to be more universal than in LCDD or LHCDD. ^, Cases of MIDD with nodular sclerosis often show mild mesangial hypercellularity and some segmental membranoproliferative features ( Fig. 9.1 ). Less common is the finding of cellular crescent formation, seen in 9% of MIDD in one series. Crescent formation (also called extracapillary proliferation ) was observed in two of three cases of alpha-HCDD. The deposits of MIDD are periodic-acid Schiff-positive and can cause ribbon-like thickening of tubular basement membranes and vessel wall myocyte basement membranes, but glomerular basement membrane thickening is uncommon. ^, Congo red staining is negative in pure forms of MIDD, although concurrent amyloidosis was described in three of 23 patients in one series. Light chain cast nephropathy is also seen in a subset of patients with MIDD. In the series by Lin et al., 11 of 34 cases of MIDD were accompanied by cast nephropathy. Among the 11 patients with MIDD and cast nephropathy, only two showed the characteristic nodular sclerosis of MIDD, two showed mild mesangial sclerosis, and the remaining seven had normal appearing glomeruli.

Light microscopic findings of monoclonal immunoglobulin deposition disease (MIDD). Nodular sclerosis is the most common appearance of glomeruli in patients with MIDD and is encountered in approximately 60% of light chain deposition disease and virtually all heavy chain deposition disease. A. (Hematoxylin and Eosin, ×400 original magnification) highlights the mesangial nodules, along with the mesangial hypercellularity that is often seen. Periodic-acid Schiff (PAS) staining ( B. ×400 original magnification) highlights nodular mesangial expansion by PAS-positive material. In contrast to diabetic nephropathy, where glomerular basement membranes are usually quite thick, glomerular basement membranes in MIDD are usually normal in thickness. Segmental double contours of the glomerular basement membrane (membranoproliferative features) are also highlighted with PAS staining.

Immunofluorescence is critical for the diagnosis of MIDD. Linear basement membrane staining for a monotypic light and/or heavy chain is a central criterion for the diagnosis of MIDD ( Fig. 9.2 ). Staining of tubular basement membranes (TBMs) appears to be essentially universal with the vast majority of cases also showing linear deposits in the glomerular basement membrane (GBM) deposits, myocyte basement membrane, vessel walls, and mesangium. ^{, ,} Immunofluorescence allows for the subclassification of MIDD cases into LCDD, LHCDD, and HCDD, depending on the staining profile of the monoclonal protein. In cases of LCDD, kappa staining predominates over lambda staining, with series showing between 68% and 91% kappa LCDD. ^{, ,} The largest series of HCDD to date shows that most HCDD is gamma-HCDD, which accounted for 12 of 15 cases in the series. The remaining cases were composed of alpha-HCDD. Cases of HCDD caused by IgM and IgD heavy chains have also been reported. Notably, HCDD cases composed of IgG3 or IgG1 heavy chains also often have C3 deposition detected by immunofluorescence, correlating with the established ability of IgG subtypes 1 and 3 to effectively fix complement.

Immunofluorescence staining in monoclonal immunoglobulin deposition disease (MIDD). All renal basement membranes can show the characteristic linear deposits of MIDD. A. (×400 original magnification) highlights strong glomerular basement membrane staining for kappa in a patient with light chain deposition disease (LCDD). Notably, prominent nodular sclerosis and mesangial deposits are not seen in this case, a finding which is not uncommon in LCDD, but would be uncharacteristic of heavy chain deposition disease. Note that Bowman’s capsule and the surrounding tubular basement membranes are also showing strong linear staining. B. (×400 original magnification) highlights the strong tubular basement membrane linear staining, that is a nearly universal feature of MIDD. A vessel is also pictured, and linear staining of the vessel wall myocyte basement membranes is also prominent in this image.

Electron microscopy of MIDD cases reveals highly electron dense, punctate, powdery deposits within renal basement membranes ( Fig. 9.3 ). TBM deposits are typically located along the outer aspect of the TBM, whereas GBM deposits are concentrated along the inner aspect of the lamina densa. Granular mesangial deposits may also be seen, a finding more commonly seen in HCDD than LCDD. ^, Lin et al. described a subset of cases where LCDD and cast nephropathy were coexistent by immunofluorescence, but the characteristic electron dense deposits of MIDD were not present. These cases were labeled as LCDD “by immunofluorescence only” and were hypothesized to represent nonspecific trapping of the monoclonal light chain in basement membranes, reflecting high circulating serum levels. The series reported by Nasr et al. required the presence of characteristic deposits by electron microscopy as one of the two diagnostic criteria, along with monoclonal protein deposition along GBMs and TBMs in immunofluorescence.

Ultrastructural features of monoclonal immunoglobulin deposition disease (MIDD). A. (×5800 original magnification) shows the powdery, punctate electron dense deposits of MIDD involving glomerular basement membranes. Within glomeruli, the deposits are usually most prominent along the subendothelial aspect of the glomerular basement membrane. B. (×5800 original magnification) shows a tubular basement membrane with powdery electron dense deposits. Note that in tubular basement membranes, the deposits are most prominent along the outer aspect of the membrane, closest to the interstitium.

Pathogenesis of monoclonal immunoglobulin deposition disease

As is true with many forms of dysproteinemia related renal disease, the disease phenotype depends in large part of the physiochemical properties of the deposited monoclonal protein. Although LCDD, LHCDD, and HCDD share many morphologic findings, such as nodular sclerosis, basement membrane thickening, and powdery deposits in renal basement membranes, there are distinct structural characteristics that promote deposition of light chains and heavy chains.

In LCDD, approximately 80% of the cases are caused by kappa light chains, which is notably different than the lambda predominance seen in light chain amyloidosis. Congé et al. published the first complete primary structure of a light chain causing LCDD. Notable findings included an overall normal kappa light chain structure with a variable domain composed of the VkappaIV gene along with the Jkappa1 . The VkappaIV sequence had several point mutations, including one resulting in an N-glycosylation site. Protein glycosylation and the development of diabetic nephropathy, which shows strikingly similar nodular sclerosis and basement membrane thickening to LCDD, has been recently investigated in a rat model of diabetic nephropathy. It is possible that the N-glycosylation in some kappa chains that cause LCDD could promote their deposition in basement membranes and cause nodular sclerosis in a similar fashion to what is seen in diabetic nephropathy. Although the VkappaIV variable region subgroup is found in approximately 7% of monoclonal kappa light chains, it appears to be disproportionately overrepresented in LCDD. Study of additional primary structures of cases of LCDD revealed multiple hydrophobic residues exposed at the molecule surface, including along complementarity determining regions (CDR). The presence of surface hydrophobic residues can have a significant influence on light chain conformation. The location of hydrophobic residues along the CDR, which is the portion of the light chain molecule responsible for contacting an antigen, suggests that there may be a specific tropism for extracellular structures, such as components of basement membranes, which appear “antigen-like” to these light chains. ^, A mouse model, in which a human LCDD VkappaIV chain was secreted, showed that the amino acid changes in the V region were sufficient to produce deposits that were similar in distribution to those observed in patients with LCDD. The increase in hydrophobicity of the light chains in LCDD may favor precipitation into tissue, because they would be less soluble in an aqueous environment. The combination of glycosylation and hydrophobicity appear to be important to the pathogenic potential of the light chains in LCDD.

In HCDD, the overarching structural abnormality found in the heavy chain is the deletion of the first constant domain CH1. Moulin et al. demonstrated CH1 deletion in four cases of gamma HCDD. The CH1 domain is essential for nascent heavy chains to interact with heavy chain binding protein in the endoplasmic reticulum. When this interaction is prevented, incompletely assembled heavy chains are prematurely secreted, without an associated light chain. Although CH1 deletion appears to be necessary for HCDD, by itself it is not sufficient. The variable regions in HCDD heavy chains commonly show mutated amino acid residues in the CDR and framework regions that result in formation of large hydrophobic regions and a positive charge at physiologic pH. This combination of charge and hydrophobicity is hypothesized to favor tissue deposition in the anionic sites of renal basement membranes. This is in comparison with heavy chain disease (HCD), which is characterized by circulating heavy chains without accompanying light chains. In contrast to HCDD, the heavy chains in HCD are not deposited in tissues. Similar to HCDD, these heavy chains frequently show CH1 deletions, allowing them to be prematurely secreted in a truncated form. In contrast, however, the variable region of the heavy chains in HCD are either fully or partially deleted, suggesting that an intact variable region is essential for tissue deposition.

Structural alterations of the light and heavy chains described earlier help explain their deposition in renal basement membranes, but the development of nodular sclerosis requires more than simple deposition of a monoclonal protein. The nodules in MIDD are largely composed of the same constituents encountered in nodular diabetic glomerulosclerosis, including type IV collagen, fibronectin, laminin, and heparin sulfate proteoglycan. The large oligomeric protein tenascin has also been found to be an essential component of the matrix accumulation in MIDD. In the normal kidney, tenascin expression is limited to the mesangial matrix, but expression becomes more widespread in a variety of pathologic conditions. An in vitro model of mesangial cells, grown in three-dimensional matrices, and incubated with light chains from patients with LCDD, showed a significant increase in tenascin-C expression, centrally located within the newly formed mesangial nodules. These in vitro experiments also showed that transforming growth factor-β (TGF-β) overexpression results in an increased production of extracellular matrix components and that platelet derived growth factor-β is largely responsible for the proliferation of mesangial cells.

Treatment and outcomes

Treatment of plasma cell dyscrasias has changed significantly since the first reported cases of MIDD, thanks to advancements in treatment for MM and amyloidosis. The primary goal of therapy is to decrease the production of the monoclonal Ig to prevent ongoing paraprotein deposition in tissue and stabilize or improve the function of affected organs. Similar to the many other rare diseases which lack large studies and randomized clinical trials, there is no standardized treatment for MIDD. Introduction of proteasome inhibitors (PI), and autologous stem cell transplantation in the treatment of MM, amyloidosis and subsequently MIDD, has dramatically changed outcomes.

If left untreated at presentation, patients who present with compromised renal function will almost invariably progress to end-stage kidney disease (ESKD) with progression taking as little as 2 months or up to 2 years in the series by Ganeval et al., in which long-term follow-up was available. In a more recent series by Li et al., all seven patients with MIDD who presented with renal disease and did not receive treatment for their underlying hematologic disorder progressed to ESKD. Alkylating agents (mainly melphalan) and prednisone were advocated as initial therapies in early series. ^, Using intermittent administration of melphalan and prednisone, a series by Heilman et al. showed overall 5-year survival and renal survival rates of 70% and 37%, respectively. This is in contrast to untreated patients who show inevitable progression to ESKD. In addition, there was improvement in proteinuria in approximately one-third of patients receiving melphalan and prednisone and renal function either stabilized or improved in approximately two-thirds of patients with serum creatinines of less than 4.0 mg/dL. In patients with serum creatinine over 4 mg/dL, over 80% progressed to ESKD despite therapy. Duration of therapy in this series ranged from 3 months to 42 months and the best results were seen in the nine patients out of 19 who received more than a year of treatment. Results in an early study by Pozzi et al. showed lower rate of response, albeit this case series had more aggressive hematologic malignancies with 11 of 19 patients meeting criteria for MM (vs. five of 19 patients with MM in the Heilman series). ^, In the Pozzi series, almost 80% of patients died within 37 months of observation despite all of the patients but one being treated with high-dose steroids and cytotoxic medications. Those with concomitant MM had worse outcomes. Plasma exchange has been used as coadjuvant, but there is no consensus regarding its benefit and its usage has declined, perhaps driven by current recommendations on management in patients with MM. In the early Pozzi series, plasma exchange did not appear to show benefits in the seven patients treated with plasma exchange and chemotherapy, because six of seven died within the first year. In a later study from the same group, the use of chemotherapy plus plasma exchange was found to be a poor prognostic factor in terms of renal and overall survival.

Recently, there have been major changes in our approach to treating MM, amyloidosis, and subsequently MIDD. Success using stem cell transplantation along with chemotherapy was reported in sporadic case reports, but a group from Italy reported five MIDD patients treated with stem cell transplantation and only one reached symptomatic renal failure during a median follow-up of 44 months; none of the five patients died over that period. In 2004 Royer et al. published a retrospective study of 11 patients with LCDD or LHCDD in which patients were treated with high-dose therapy with the support of autologous stem cell transplant. The protocol for mobilization of stem cells and the subsequent therapeutic protocol varied somewhat in the 11 patients based on when they were treated (before and after 1995) but in all cases, high-dose therapy was supported by the reinfusion of at least 2 × 10 ⁶ /kg CD34-positive cells. In eight of 10 patients who initially had a monoclonal gammopathy, high-dose therapy and autotransplantation produced a decrease in the level of monoclonal immunoglobulin. All three patients with nephrotic syndrome experienced remission. Within a median follow-up of 51 months, three patients were re-treated because of MM relapse; one died 93 months posttransplant because of progressive myeloma and one required hemodialysis. Weichman et al., also reported encouraging results with high-dose melphalan and autologous stem cell transplant. In five cases with LCDD, during a median follow-up of 12 months (range 4–29 months), all patients remained alive and well. Complete hematologic response was achieved in 83% of the cases, and the median percentage reduction in proteinuria was 75.3% (range 38.7–89.3). All patients had normalization of their serum free light chain assays. In addition, there were no major side effects (Grade 4 toxicities) related to the therapy. Firkin et al. described a patient with LCDD and incipient dialysis dependency who showed sustained improvement in renal function resulting in discontinuation of dialysis after myeloablative melphalan and autologous stem cell transplant. This improvement in renal function and proteinuria was noted despite a repeat renal biopsy performed at 7 months after transplantation that did not show any diminution of the kappa light chain deposits. At 20 months after transplant, creatinine clearance had increased from 14 mL/min to 44 mL/min. Series by Hassoun et al. and Lorenz et al. provided additional experience confirming the efficacy of high-dose chemotherapy and autologous stem cell transplant, and in both of these series, kidney transplantation was performed in a patient with ESKD who had achieved hematologic remission, with good follow-up results. With respect to potential toxicity of these regimens, one of these series reported the death of one patient caused by multiorgan failure at 26 days postautologous stem cell transplant. Renal transplantation, although a viable consideration in patients with ESKD who have achieved hematologic remission, should not be considered in patients without the elimination of light chain production based on the experience at the Manchester Royal Infirmary and the Mayo Clinic. ^,

More recently, bortezomib is gaining an increased role in the treatment of MIDD. Bortezomib has multiple mechanisms with the potential to improve outcomes in patients with MIDD. Bortezomib both suppresses production of the responsible paraprotein and also inhibits the nuclear factor kappa-light-chain-enhancer of activated B cells pathway that appears to be involved in mesangial cell proliferation and mesangial sclerosis. Bortezomib has also been shown to decrease levels of TGF-β1 levels, which are profibrotic and have been implicated in the pathogenesis of nodule formation in MIDD. Bortezomib may also have a role in downregulating collagen and tissue inhibitor of metalloproteinases 1 production. Thus bortezomib may decrease the progression of glomerulosclerosis, improve glomerular filtration rate, and reduce proteinuria by interrupting this cascade of events. Bortezomib along with dexamethasone was shown to have excellent results in four patients in whom two of them were treatment naïve. There was complete hematologic response in two patients, whereas the others had a decrement in their free light chains of greater than 50%. Three of these patients later underwent autologous stem cell transplant and were disease free after 10 to 18 months of follow-up. Another group reported on three patients with LCDD induced with bortezomib in combination with cyclophosphamide and dexamethasone. Because of severe side effects, the authors recommended a less aggressive approach with patients with LCDD than patients with MM. In patients with recurrence of MIDD after renal transplantation, bortezomib (plus dexamethasone or lenalidomide and dexamethasone) has shown the ability to salvage the allograft with excellent long-term results, including continuing function for at least 10 years. In patients with HCDD, bortezomib-based therapy seems to show encouraging results, with a recent series showing 70% of patients in reaching hematologic response and 90% with renal response.

There is still significant variability in the treatments chosen for patients with MIDD, specifically in patients with LCDD as shown in the recent report by Sayed et al., in which multiple cytotoxic and immunomodulatory agents were used (thalidomide, lenalidomide, bortezomib, alkylator based, melphalan plus autologous stem cell transplant, and steroids alone). Nonetheless, complete remission (CR) was achieved more frequently in patients treated with bortezomib. Similar success with bortezomib was reported in series by Cohen et al. and Kourelis et al. Regimens included bortezomib plus dexamethasone; bortezomib plus lenalidomide plus dexamethasone; bortezomib plus thalidomide plus dexamethasone; cyclophosphamide plus bortezomib plus dexamethasone; and high-dose melphalan followed by autologous stem cell transplant. ^, In the largest series to date with MIDD treated with bortezomib (49 patients of which 35 had LCDD; 12 HCDD; and 12 LHCDD), 25 were only treated with bortezomib. The overall hematologic response rate was 91%. With a median follow-up of 54 months, of 38 patients treated with bortezomib as first-line of therapy, only one had renal relapse after almost 53 months. Of the total cases, 53% of them achieved renal response with estimated glomerular filtration rate (eGFR) improvement of 35% and proteinuria decreasing from a median of 1.5 to 0.2 g/day. Similarly, in the largest case series of patients with MIDD described, patients receiving autologous stem cell transplant or PI-based therapies were more likely to achieve CR or very good partial remission (VGPR) compared with those receiving other therapies (66% vs. 2%, p < .0001). In those patients achieving CR/VGPR, renal survival at 5 years was 77% compared with 54% of those achieving partial response (PR) or no response.

Autologous stem cell transplant has a peritransplant mortality (within the first 100 days of engraftment) of less than 10%. ^, In the study by Sayed et al., among 11 patients not requiring dialysis, median eGFR before autologous stem cell transplant was 24 mL/min, which had increased among the 10 surviving patients to 38 mL/min by end of follow-up. Thirteen of 15 patients who survived autologous stem cell transplant achieved a hematologic CR and two achieved PR. Patients with CKD 5 not on dialysis did have worse outcomes. In the study by Kourelis et al., overall outcomes were better in patients who underwent autologous stem cell transplant compared with those with PI and other treatments and were more likely to achieve at least a CR/VGPR: 77% for autologous stem cell transplant versus 56% for PI-based therapies versus 6% for other treatments, ( p < .0001). Patients who achieved hematologic CR were more likely to achieve renal response as well. In a retrospective study of four patients with MIDD on dialysis, high-dose melphalan and autologous stem cell transplant were analyzed. All of them underwent kidney transplant at a median of 2.6 years after the stem cell transplant and all patients were alive by the time of publishing in 2018. It was suggested by the authors that this therapy was feasible and toxicity and mortality seemed to be acceptable even in selected patients with ESKD secondary to MIDD. The treatment of patients with MIDD should therefore have a multidisciplinary approach with both nephrologists and hematologists involved.

As presented previously, outcomes in MIDD are dependent to a large extent on the therapy provided. One subset of MIDD patients that merits specific mention is those who present with MIDD and coexisting light chain cast nephropathy. Patients with coexisting cast nephropathy have a significantly worse prognosis, both in terms of progression to ESKD and overall patient survival, compared with those without cast nephropathy. ^, When comparing outcomes, especially when using some of the early literature, effort needs to be made to separate cases of “pure” MIDD from cases of MIDD with cast nephropathy. In general, patients with severe cardiac disease have worse prognosis as well. Patients with pulmonary presentation, in whom a nodular pattern is observed, tend to have better prognosis compared with those with diffuse pulmonary LCDD. Low hemoglobin at presentation is also associated with worse ESKD prognosis and worse survival, and patients who present with more advanced renal failure tend to have a less robust renal response to therapy. ^,