Radioimmunodetection of amyloid deposits in patients with AL amyloidosis

Light chain–associated (AL) amyloidosis is a monoclonal plasma cell dyscrasia characterized by the pathologic deposition in vital tissues of fibrils formed from κ or λ immunoglobulin (Ig) light chain–related components.^1-3  The relentless accumulation of such fibrillar material typically leads to progressive organ dysfunction and death within 18 to 36 months. In the case of cardiac involvement, the prognosis is even more ominous, with a survival time of 3 to 9 months; fewer than 5% of all AL amyloidosis patients live more than 10 years after diagnosis.⁴  Currently, therapeutic options are limited to diminishing light chain production with anti–plasma cell chemotherapy (eg, melphalan and/or corticosteroids) given in conventional amounts or high doses combined with autologous stem cell transplantation.^4-9  This approach, which is based on the premise that reduction in synthesis of the amyloidogenic precursor will slow fibril formation, has extended length of life and, in some instances, resulted in improved organ function over time; nonetheless, the prognosis remains poor because of persistent amyloid burden.

To address this issue, we have focused on passive immunotherapy as a means to expedite removal of amyloid deposits and, through these research efforts, developed a murine (m) IgG1 anti–human light chain monoclonal antibody (mAb), designated 11-1F4, which recognized a conformational epitope present on amyloid fibrils, but not the soluble amyloidogenic precursor protein.^10,11  Furthermore, when administered to mice bearing subcutaneous human AL amyloidomas, the antibody bound to the pathologic material and initiated an inflammatory response that led to elimination of the induced tumors.¹²  Notably, we also demonstrated that m11-1F4, after radiolabeling with the positron-emitting isotope I-124, imaged the xenograft, as evidenced by micro–positron emission tomography/computed tomography (PET/CT).¹³  These results have led to a Food and Drug Administration (FDA)–approved Phase I Exploratory investigational new drug (IND 100472) study to determine the safety and biodistribution of ¹²⁴I-m11-1F4 in patients with AL amyloidosis. We now report the results of this trial that have involved, to date, 18 subjects in whom the radioiodinated antibody was well tolerated, elicited no human anti–mouse antibody (HAMA) response, and notably in 9 subjects, was taken up by organs deemed to contain amyloid.

All 18 patients entered on study (Table 1)¹⁴  were HAMA-negative and had AL amyloidosis based on accepted clinical and laboratory criteria,¹⁵  as well as (with 1 exception) the results of chemical analysis¹⁶  of amyloid extracted from tissue or fat biopsy specimens. All subjects provided written informed consent in accordance with the Declaration of Helsinki under a protocol approved by the FDA and the University of Tennessee Graduate School of Medicine Institutional Review Board.

Good medical practice-grade m11-1F4 (National Service Center No. 740550) and isotope I-124 were furnished by the National Cancer Institute–Frederick Cancer Research and Development Center's Biologic Resource Branch and by IBA Molecular, respectively. The antibody was radioiodinated using Iodogen (Pierce) as an oxidant, purified by solid phase size-exclusion chromatography (PD-10 desalting column; GE Healthcare), and eluted with sterile phosphate-buffered saline. The peak protein–bound radioactive fraction was diluted to 3 mL with phosphate-buffered saline, passaged through a 0.22-μM pore-sized filter, and an aliquot removed to determine protein concentration, radiochemical purity, specific activity, stability, and immunoreactivity. The final product, which was tested for sterility and endotoxin content, was prepared for patient injection by further dilution to 30 mL and contained less than 1 mg of m11-1F4 labeled with approximately 2 mCi (74 MBq) of I-124, plus 5% human serum albumin and 15 mg of ascorbic acid. To inhibit thyroidal uptake of free radioiodide, 0.3 mL of super saturated potassium iodide was prescribed 24 hours before ¹²⁴I-m11-1F4 administration (and then continued for an additional 9 days); the next day, patients were premedicated with 25 mg of diphenhydramine/650 mg of acetaminophen and infused over approximately 15 minutes with the reagent.

PET/CT images were obtained using a Siemens Biograph mCT instrument that consisted of a low-dose CT from midthigh to crown, followed by a series of seven 5-minute PET acquisition scans covering the same region. After online correction for random scatter, prompt gamma emission, and attenuation (CT-based correction algorithm), these data were reconstructed by a point spread function-based iterative algorithm (TrueX; Siemens) with an image matrix of 168 × 168 that provided an approximately 8-mm full-width half-maximum resolution. Maximum intensity projection images were generated using Siemens Inveon Research Workplace software (Version 3.0).

Six-μM thick sections of formalin-fixed, paraffin-embedded tissue were subjected to antigen retrieval using Citra Plus or Glyca (BioGenex) and incubated at 4°C overnight with 1 μg/mL mAb m11-1F4, followed by biotinylated goat anti–mouse IgG, and then the avidin-biotin complex solution (ABC; Vector Laboratories). Consecutive tissue sections also were stained with Congo red. Slides were examined by light and polarizing microscopy.

The AL fibril reactivity of the m11-1F4 mAb was not affected by radioiodination and, in all cases, the administered preparations were sterile, had negligible endotoxin content, were well tolerated, and elicited no HAMA response in serum specimens obtained 60 days later. For dosimetry purposes, the first 3 patients were scanned 3, 5, 48, 72, 120, and 168 hours after infusion. The radiolabeled antibody plasma t_½ was approximately 25 hours, which was longer than that seen in mice,¹³  but consistent with clearance of mIgG in humans (ie, ∼ 30 hours).¹⁷  The calculated effective radiation dose (0.4 mSv/MBq) proved acceptable to the FDA. By 48 hours, approximately 70% of radioactivity in the blood pool had cleared (although, in the case of patient AL 2, amyloid-associated binding of mAb 11-1F4 in the mediastinal lymph nodes persisted for 168 hours, with a mean activity of 2.1 MBq/mL). In subsequent studies, subjects were imaged only on days 2 and 5 after infusion.

The results from the 18 patients are provided in Table 1. In 9, the PET/CT scans revealed uptake of the radiolabeled antibody in areas deemed to contain amyloid (eg, liver, lymph nodes, bone marrow, and intestine, as well as spleen, which may represent another source of the amyloidogenic precursor protein).¹⁸  In contrast, those with cardiac or renal amyloid had no demonstrable uptake in these sites. In 3 of 5 subjects with positive liver imaging, the serum alkaline phosphatase concentrations were abnormally high. There was no evident correlation between the radioimmunoimaging data and disease duration or therapy.

Because the immunoreactivity of m11-1F4 was not affected by labeling with I-124, we investigated whether the in vivo results could be related to those derived immunohistochemically using diagnostic tissue biopsy specimens available from 14 of the 18 cases. In these studies, which used the unmodified antibody, the reagent immunostained (Figure 1) the deposits in 10 specimens, of which 6 had positive PET/CT scans, and in 4 of 6 specimens that did not image.

We previously had shown through peptide mapping that the specificity of mAb 11-1F4 depends upon a conformational epitope present on light chain fibrils that is not exposed on the native protein.^10,11  Thus, the inability of the radiolabeled antibody to bind certain AL deposits, both in vivo and in vitro, may reflect a structural alteration in this cryptic epitope or its inaccessibility, as seen in the cases of renal amyloid in which (in contrast to other tissues) this material was immunostained weakly, if at all, by the reagent. Alternatively, the concentration of the immune target per unit volume may have been too low and therefore undetectable by PET imaging. In the 3 cases in which the amyloid was immunostained by mAb 11-1F4 but was not imaged, it is possible that a higher dose of radiotracer would have yielded a positive result. Of note, although radiolabeled serum amyloid P component can be used to visualize renal deposits, it also is incapable of imaging cardiac amyloid,¹⁹  presumably because of vascular factors.

Given these results, we posit that ¹²⁴I-m11-1F4 radioimmunoimaging could be used to predict which AL patients would be candidates for passive immunotherapy using the chimeric version²⁰  of this mAb, which currently is under production for an eventual Phase I clinical trial. Notably, the modified amyloid-reactive antibody, in contrast to the murine form, advantageously, could be administered repeatedly and has a considerably longer t_½. This novel approach, namely passive immunotherapy, would offer an additional therapeutic option for patients with this invariably fatal disorder.²¹ 

The assistance over the course of the study of W. Thompson, R. Geldrich, T. Williams, S. Macy, and C. Wooliver is gratefully acknowledged.

This work was supported in part by Orphan Grant FD-R-003420 from the FDA, US Public Health Service Research Grant CA10056 from the National Cancer Institute/National Institute of Diabetes and Digestive and Kidney Diseases, and the Aslan Foundation. A.S. is an American Cancer Society Clinical Research Professor.

National Institutes of Health

Contribution: J.S.W., S.J.K., D.W.T., and A.S. designed the study; S.J.K. radioiodinated the mAb and J.S.W. and S.J.K. performed quality control assays for the product; A.C.S. and M.J.L. were responsible for the PET/CT imaging and data processing; G.T.S., K.J.W., and Y.F. reviewed the PET/CT images; M.G.S. provided the dosimetry data; and J.S.W., D.T.W., and A.S. wrote the paper.

Conflict-of-interest disclosure: J.S.W. and A.S. have intellectual property rights for the use of mAb 11-1F4 in the diagnosis and treatment of patients with amyloidosis. D.W.T. is a consultant for Siemens. The remaining authors declare no competing financial interests.

The current affiliation for D.W.T. is PET and SPECT Development Group, Singapore Bioimaging Consortium, Singapore.

Correspondence: Jonathan S. Wall, Department of Medicine, Human Immunology and Cancer Program, University of Tennessee Graduate School of Medicine, 1924 Alcoa Hwy, Knoxville, TN 37920; e-mail: jwall@utmck.edu.