Prognostic Value of Dynamic Monoclonal Protein Pattern on Probability of Myeloma Progression from the Precursor State

Multiple Myeloma (MM) is a cancer of terminally differentiated plasma cell resides in bone marrow, which is always preceded by clinically asymptomatic precursor states. The process of malignant transformation however is not fully understood. Analyses of cells from precursor state have provided evidence that it is a genetically advanced lesion, wherein tumor cells carry many of the genetic changes found in MM cells. Furthermore, mice xenografts from patient (pt) with precursor disease showed progressive growth on its own suggesting progression potential is counteracted by active extrinsic restraints that prevent its progression to MM (Das R et al. Nature Medicine, 2016). Hence, precursor progression to MM can be viewed as a dynamic process in which the clonal mass is the net result of pro-growth, pro-survival replicative capacity of the myeloma clones vs. anti-tumor immunity. The change in the net clonal mass sometimes is large enough to be detected by measurable fluctuations in serum monoclonal protein spike (M-spike) levels and reflects a highly dynamic battle between plasma cell clone and cell-mediated anti-myeloma immunity among those pts. The prognostic value of this highly dynamic state on risk of progression to MM is largely unknown. In this study we sought to assess the association between the dynamic M-spike pattern and the risk of progression from a precursor clone to MM. Methods: All pts with positive serum electrophoresis (SPEP) values measured from January 2011 to January 2021 for our institution were included for this study. All pts that had <3 M-spike readings before requiring anti-myeloma regimen, received active treatment for MM within 90 days of the first positive SPEP, had M-spike values >3g/l or had clear evidence of progression by serial increasing M-spike values were excluded.. We assigned the pattern of M-spike values over time into two groups according to stability. To eliminate the test-to-test variability, we defined stable disease as lack of more than 20% difference from lowest M-spike, and fluctuating disease if the difference exceeds 20%. The time-to-treatment (TTT) was then measured from the first positive SPEP to the time of treatment and was censored at the date of last follow-up for pts who did not have anti-myeloma treatment with death as competing risk. The effect of M-spike pattern on TTT, treating as binary outcome, was also evaluated using logistic regression analysis. Univariate analysis of TTT was performed using Gray's method. All tests were two-sided and p-value ≤ 0.05 were considered statistically significant. Results: A total of 565 pts were studied, 193 with a fluctuating pattern and 372 with a stable pattern. Median age at first positive SPEP was 76 years old (range: 38-101), 271 (48%) of them were male; 427 (75%) were Caucasian and 123 (22%) African American. Baseline creatinine was 1.17 mg/dl (range 0.5-11.2 SD 0.94). Baseline albumin was 3.97 g/dl (range 1.6-5.2 SD 0.47), 336 pts (59%) were IgG, 75 (13%) IgA, and 142 (25%) had IgM M-spike; 326 (58%) pts had Kappa light chain and 239 (42%) had Lambda. Autoimmune diseases were more common in the stable disease pattern (17% vs. 10% p=0.028), the most frequent one was Rheumatoid arthritis in 33 pts (6%). Furthermore, baseline albumin was statistically higher in pts in the stable cohort (4.04 vs. 3.85g/dl, p=<0.0001). Median follow up was 46.9 months (range: 3.2-120.9). Logistic regression: the odds ratio (OR) of having progression to treatment for fluctuation group was 4.5 (95% CI: 1.37-14.18, p= 0.013) in comparison to stable group. Time-to-event analysis: pts with fluctuating M-spike had higher chance of needing therapy (Hazard Ratio: 3.73, 95% CI: 1.19-11.66, p=0.024) than stable group. Cumulative incidence of needing treatment is shown in Figure-1. Conclusion: The dynamic pattern of M-spike seen here with fluctuating values over time (over time is two words) was significantly predictive of progression to MM in pts with a precursor disease state. These findings may reflect a different nature of tumor vs. immune system balance and possible impaired immunity in these pts which may have implications for future clinical trials that evaluate the efficacy of anti-myeloma immunotherapeutics. Furthermore, our finding suggest that fluctuations in M-spike levels may be factored into risk models that predict the progression to MM among pts with precursor clones. Larger studies are warranted to further study this concept. Figure 1 Figure 1. Disclosures Malek: Medpacto Inc.: Research Funding; Janssen: Other: Advisory board ; Takeda: Honoraria; Cumberland Inc.: Research Funding; Amgen: Honoraria; BMS: Honoraria, Research Funding; Bluespark Inc.: Research Funding; Sanofi: Other: Advisory Board.

High Prevalence of Monoclonal Gammopathy in a Population at Risk: The First Results of the Promise Study

Background Multiple myeloma (MM) evolves from monoclonal gammopathy of undetermined significance (MGUS), a clinically detectable but asymptomatic premalignant phase seen in ~3% of the general population 50 years of age or older. The prevalence of MGUS has not been described in a population at high risk of developing MM, specifically Black/African American (AA) individuals or first-degree relatives of patients with hematologic malignancies (HM). In 2019, we launched the first nationwide US screening study for individuals at high risk of MM to help better identify what population would benefit most from screening and early intervention for precursor MM stages. We aim to assess the prevalence of MGUS in a population at high risk of MM and characterize clinical variables of individuals who screen positive. Here, we report interim screening data on the first 2,960 participants. Methods Individuals aged 40 or older with an additional MM risk factor are eligible to be screened in the PROMISE Study. High-risk individuals include Black/AAs and those with a first-degree relative diagnosed with a hematologic malignancy or a precursor condition to MM. Blood from all participants was analyzed via serum protein electrophoresis, immunofixation, and Optilite® to measure the serum free light chains (sFLC), IgG, IgA and IgM. Results were returned to all participants, and those who tested positive for a monoclonal gammopathy (MGUS/SMM) were referred to a hematologist for clinical follow-up and invited to periodically complete epidemiologic exposure and psychosocial questionaries, including a 4-item cancer worry questionnaire and the RAND 36-item Short Form Survey (SF-36). To investigate the use of the higher-sensitivity matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) with the Optilite® IgG, IgA, IgM and sFLC results as a screening test for all participants, we rescreened 1,092 samples from PROMISE. The Binding Site Group proprietary software was used for the analysis of the combined MS/Optilite® results, allowing for the detection and quantification of M-protein. Heavy-Chain MGUS (HC-MGUS) was defined by the presence of one or more paired heavy and light chain monoclonal peaks detected by MS. Pairing was based on mass to charge ratio of identified peaks. To enrich the PROMISE cohort with Black/AA individuals, we identified and screened 1,868 Black/AA additional individuals from the Mass General Brigham (MGB) Biobank who met the PROMISE enrollment criteria. Screening was performed by MS/Optilite®, and results have not been returned to participants. Results We screened 2,960 participants with the combined MS/Optilite® approach. We report here the prevalence of HC-MGUS and plan on presenting the estimated rate of light chain MGUS in our cohort, at the meeting. We detected HC-MGUS in 9.6% (95% CI: 8.6-11%) of our cohort, with a prevalence of 10% (95% CI: 8.3-12%) in the PROMISE cohort and 9.4% (95% CI: 8.1-11%) in the MGB cohort (Table 1 and Figure 1). HC-MGUS prevalence increased with age in high-risk individuals from 4.9% (CI: 3.3-6.9%) for participants aged 40-49 to 13% (CI: 10-17%) in the 70-79 range (P < 1.2E-5). Among monoclonal HC-MGUS, we found 65% IgG, 18% IgM, and 18% IgA. M-spike was quantified in 97% of samples. Median M-spike concentration was, 0.058g/dL (max. 2.6g/dL) for IgG, 0.0043g/dL (max. 0.6g/dL) for IgM, and 0.067g/dL (max. 0.8g/dL) for IgA. In the Promise cohort, no significant change in cancer worry was observed across the pre- and post-screening interval among participants who screened positive (P = 0.52). Health-related quality of life, as measured by the SF-36, was not significantly different in screen-positive vs. screen-negative individuals for any of the eight subscales (all P > 0.20). Conclusions We present the largest dataset on monoclonal gammopathy prevalence and screening in individuals at high risk for MM, and more specifically the largest cohort of Black/AA, using a novel high-sensitivity testing approach. Our results confirm that older adults who are Black/AA or have a first-degree relative with an HM have a high prevalence MGUS and may benefit from precision screening approaches to allow for early detection and clinical intervention. Preliminary data on cancer worry and quality of life indicates that the psychosocial burden of screening in this population is likely minimal. Figure 1 Figure 1. Disclosures Sakrikar: The Binding Site: Current Employment. Krause: The Binding Site: Current Employment. Barnidge: The Binding Site: Current Employment. Bustoros: Takeda: Consultancy, Honoraria; Janssen, Bristol Myers Squibb: Honoraria, Speakers Bureau. Perkins: The Binding Site: Current Employment. Harding: The Binding Site: Current Employment, Membership on an entity's Board of Directors or advisory committees, Patents & Royalties. Kapoor: Ichnos Sciences: Research Funding; Amgen: Research Funding; Pharmacyclics: Consultancy; Takeda: Research Funding; Sanofi: Consultancy; Regeneron Pharmaceuticals: Research Funding; AbbVie: Research Funding; BeiGene: Consultancy; Sanofi: Research Funding; Karyopharm: Research Funding; Glaxo SmithKline: Research Funding; Karyopharm: Consultancy; Cellectar: Consultancy. Mo: Sanofi: Honoraria, Membership on an entity's Board of Directors or advisory committees; GSK: Consultancy, Membership on an entity's Board of Directors or advisory committees; BMS: Membership on an entity's Board of Directors or advisory committees; Epizyme: Consultancy; Eli Lilly: Consultancy; Janssen: Honoraria; Karyopharm: Honoraria, Membership on an entity's Board of Directors or advisory committees. Murray: The Binding Site: Patents & Royalties: Potential Royalties for use of mass spectrometry in M-protein detection. Getz: Scorpion Therapeutics: Consultancy, Current holder of individual stocks in a privately-held company, Membership on an entity's Board of Directors or advisory committees; IBM, Pharmacyclics: Research Funding. Marinac: JBF Legal: Consultancy; GRAIL Inc: Research Funding. Ghobrial: AbbVie, Adaptive, Aptitude Health, BMS, Cellectar, Curio Science, Genetch, Janssen, Janssen Central American and Caribbean, Karyopharm, Medscape, Oncopeptides, Sanofi, Takeda, The Binding Site, GNS, GSK: Consultancy.

Association of Progression Patterns in Monoclonal Gammopathy of Undetermined Significance (MGUS) with Outcome in Multiple Myeloma: A Cohort Study of over 35 Years

Introduction: Guidelines suggest following monoclonal gammopathy of undetermined significance (MGUS) according to risk of multiple myeloma progression. Follow-up of low-risk MGUS is debated as progression risk is low. Worse myeloma-related outcome was observed in patients followed for low-risk MGUS, potentially due to less optimal follow-up (Bianchi et al. Blood 2010, Sigurdardottir et al. Jama Oncology 2015). However, it is not clear if progressing low-risk MGUS by its nature displays more aggressive biological behavior. To gain better understanding of progression patterns in MGUS, we investigated, independent of follow-up, whether progression from low-risk MGUS is associated with worse outcome in multiple myeloma. The main outcome was overall survival from the time of myeloma diagnosis. Methods: The Northern Sweden Health and Disease Study (NSHDS) is a longitudinal prospective cohort with more than 100,000 participants. Typically, NSHDS participants donate repeated blood samples at intervals of several years. Samples are frozen within one hour and stored at -80° C at Umea University Hospital. Linkage to the Swedish Cancer Registry facilitated identification of myeloma patients with two pre-diagnostic samples in NSHDS (N = 61). Of these, we screened repeated pre-diagnostic samples using protein- and immunofixation electrophoresis and free light chain assays. We identified 42 individuals who had detectable monoclonal gammopathy in both pre-diagnostic samples without MGUS follow-up before myeloma diagnosis, allowing to investigate natural progression patterns in relation to outcome. Overall survival was determined using Kaplan-Meier plots. Hazard ratios and 95% confidence intervals were calculated using multivariable Cox regression including known prognostic factors. Fisher's exact test was used to compare categorical variables. Results: The first pre-diagnostic sample was collected in November 1986 and the last follow-up since myeloma diagnosis was in February 2021, resulting in a 35-year study duration. Median age at myeloma diagnosis was 62 (range 48-84) with a median follow-up of 7 years (range 0.2-18). At first pre-diagnostic blood draw, 12 had low-risk and 30 had MGUS of other risk (Mayo Clinic criteria). Characteristics at myeloma diagnosis except sex were similarly distributed between the two groups. Comorbidities, myeloma treatment and access to novel drugs were balanced between groups. Bone disease (osteolytic lesions and/or vertebral compression fractures attributable to myeloma) at myeloma diagnosis was more common in patients who had low-risk MGUS at first pre-diagnostic blood draw (P = 0.04). This association was pronounced excluding light chain myeloma (P = 0.01). Access to other radiographic imaging than conventional skeletal survey such as CT or MRI was similar for both groups. In low-risk vs. other MGUS, median overall survival since myeloma diagnosis was 2.3 (95% CI, 1.8-2.9) years and 7.5 (95% CI, 4.8-10.2) years (Figure A). Results were similar investigating overall survival since frontline therapy start (excluding 5 patients not requiring therapy) (Figure B). Sex and bone disease were not associated with survival. At repeated pre-diagnostic blood draw (in median 3.7 years prior myeloma diagnosis), 67% vs. 19% had low- or low-intermediate risk MGUS in patients with low-risk vs. other MGUS at first pre-diagnostic blood draw (P = 0.01), suggesting more rapid progression close to myeloma diagnosis in patients with low-risk MGUS at first blood draw. Investigating this further, we plotted M spikes in low-risk vs. other MGUS of IgG isotype (Figures C-D). M spike trajectories were largely similar between groups, although the annual median M spike increase from repeated pre-diagnostic blood draw to myeloma diagnosis was 6.0 g/L in low-risk vs. 2.3 g/L in other MGUS (P = 0.14). Conclusions: Progression from low-risk MGUS is, independent of MGUS follow-up, associated with a higher proportion of bone disease and worse survival. Based on the known phenotypic heterogeneity in multiple myeloma, we speculate that low-risk MGUS in case of malignant progression belongs to a group of more aggressive tumors. Our results need to be interpreted carefully because of the small sample size. Replication and further investigation are needed. If replicated, these findings could help to improve current MGUS follow-up strategies, which are solely based on progression risk. Figure 1 Figure 1. Disclosures No relevant conflicts of interest to declare.

Eight-Color Flow Cytometry Phenotypic Markers and Disease Progression in Monoclonal Gammopathy of Unknown Significance

Introduction: Flow cytometric immunophenotyping is considered an indispensable tool for the diagnosis, classification, and monitoring of plasma cell disorders. Herein, we seek to study the clinical significance of expression of phenotype markers in monoclonal gammopathy of unknown significance (MGUS). Methods: We identified a cohort of patients with a diagnosis of MGUS from the institutional myeloma database. Bone marrow (BM) aspirate assessment was performed using 8-color immunophenotypic next-generation flow cytometric (NGF) analysis with a minimum sensitivity of 10 -5 cells at the time of diagnosis or first visit to our institution. BM aspirate samples were immunophenotyped on a FACSCanto II flow cytometer using antibodies (BD) to delineate normal and abnormal plasma cells [CD138 (V-500), CD38 (FITC), CD19 (PE-Cy7), CD45 (V-450), CD27 (PercpCy5.5), CD81 (APC-H-7), CD56 (APC) and CD20 (PE)]. The sensitivity or the Limit of Detection (LOD) for this assay was validated to 20 cells in 2 ×10 6 events (0.001%), and the reproducibility or Lower Limit of Quantitation (LLOQ) is 50 cells in 2 ×10 6 events. Clinical and laboratory variables were also collected. Based on previously published data, expression (CD19, CD45, CD81), and lack of expression (CD56, CD27, CD20) of the above-mentioned surface antigens were analyzed. Additional variables such as IgA isotype, size of M-protein (≥15 g/L), and abnormal free light chain ratio(abnFLR) (defined as <0.1 or >10) were included in regression fitting models. Results: A total of 157 patients with MGUS were included in this analysis. The median age at diagnosis was 60 years (range 24- 84), 84 (53 %) patients were female and 25 (16%) were African American. Overall, IgG Kappa (75/148, 50%) was the most common isotype. Fluorescent-in-situ hybridization (FISH) data were available in 35 patients with t (4:14) and t (14;16) seen in 3 patients each. At a median follow-up of 18.15 years (quantiles 11.35, 33.62), 28 patients experienced disease progression (25 to MM, 2 to Waldenstrom macroglobulinemia, and 1 Smoldering myeloma). The median progression-free survival of this cohort was 17.3 years. Among these, occurrences of the bone lesion (8/28; 28.6%) were the most common pattern of disease progression to MM. This analysis showed lower odds of progression with the expression of CD27 (OR-0.39, 95% CI 0.15-0.99) (figure 1A). Disease progression was more common in patients with an abnormal plasma cell clone size ≥ of 3.1% at diagnosis (60% vs. 12.5%, p=0.0005). An abnormal plasma cell clone of ≥3.1% at diagnosis, was associated with increased odds of progression (OR-10.79, 95% CI 4.02-28.98) and a shorter PFS (12.5 years versus NR, p=0.01) (figure 1B). Serum M-spike ≥1.5 g/dL (OR-3.54;95% 1.30-9.62) and abnFLR (OR-2.30, 95% CI 1.00-5.32) were also associated with a higher odds of progression. However, in this population, the presence of IgA isotype did not increase the odds of MGUS progression. In a stepwise regression model, serum M-spike≥1.5 g/dL, abnFLR, and the lack of expression of CD27 were associated with the risk of disease progression. Conclusion: In addition to previously published risk factors, our cohort shows that the expression of CD27 antigen by eight-color flow cytometry confers a lower risk of disease progression of MGUS. This is consistent with our previous report that CD27 is progressively down-regulated in the transition from normal plasma cells (NPC) to MGUS to MM (Zhan et al, Blood 2006). Furthermore, we show that size of the myeloma clone (≥ 3.1% ) is a possible surrogate marker for disease progression in MGUS. Figure 1: 1A shows forest plot of odds ratios for flow cytometry markers, IgA isotype, size of M protein, abnFLR and plasma cell clone size. 1B shows the Kaplan Meier estimates of PFS for patients stratifies by plasma cell clone size. Figure 1 Figure 1. Disclosures Mohan: Medical College of Wisconsin: Current Employment. Atrash: GSK: Research Funding; AMGEN: Research Funding; Jansen: Research Funding, Speakers Bureau.

The efficiency of hypogammaglobulinemia reflex immunofixation electrophoresis in identifying monoclonal proteins

Introduction/Objective Hypogammaglobulinemia can be a common occurrence in disorders with monoclonal gammopathies. Because hypogammaglobulinemia may mask a monoclonal protein on serum protein electrophoresis (sPEP), its presence in the absence of a discernible M-spike is often the basis of reflexive testing by immunofixation electrophoresis (IFE). At our Institution, reflex IFE has historically been performed in cases where the gamma fraction on sPEP is <0.6 g/dL. The aim of this study was to test the predictive performance of hypogammaglobulinemia in identifying abnormal bands on IFE in newly screened patients. Methods/Case Report All patients that underwent sPEP testing from November 2020 to May 2021 at our Institution were identified. Among them, patients with gamma fraction <0.6 g/dL and no previous sPEP testing were included for analysis. Reflex IFE results were reviewed for identification of abnormal bands. Results (if a Case Study enter NA) Out of a total number of 1,374 patients tested for sPEP in the study period, 72 had serum gamma fraction <0.6 g/dL (5.2%). Among them, 36 patients had no previous sPEP testing, and their reflex IFE were reviewed. In 38.8% of the cases, the IFE showed one or more abnormal (monoclonal) bands. When considering a new threshold for hypogammaglobulinemia IFE reflex of <0.4 g/dL, the diagnostic yield for finding abnormal bands increased to 62.5%. Moreover, a percentage reduction of 64.2% was observed in the number of reflex IFE performed. Conclusion Although these data must be confirmed using a larger sample population, a lower threshold for hypogammaglobulinemia may be proposed at our Institution to reduce labor and costs and to improve efficiency of monoclonal protein detection.

Diagnostic Challenge in a Patient Presenting with Ascites and Hypergammaglobulinemia

Ascites is defined as the accumulation of intra-peritoneal fluid that can be caused by several diseases. We described a 47-year-old female presenting with low serum-ascites albumin gradient (SAAG) and a markedly high level of serum globulin. Serum protein electrophoresis revealed an M spike in the gamma region. Other laboratory results showed a marked increase in aspartate aminotransferase and alanine aminotransferase and predominantly conjugated hyperbilirubinemia without a sign of dilatation of bile ducts from abdominal ultrasonography examination. Furthermore, the follow-up showed a positive result for the anti-nuclear antibody test. The patient was assessed with autoimmune hepatitis, and the cause of ascites was suggested from portal hypertension although the level of SAAG was low. The ascites condition got improved after salt restriction, diuretics treatment, and abdominal paracentesis. However, the patient passed away because of the intracranial hemorrhage as a result of prolonged INR and APTT due to liver failure.

Idiopathic Multicentric Castleman Disease

Castleman Disease (CD) is a spectrum of heterogenous hematological diseases that share characteristic clinical, histopathological, and immunological features. We present a case of a 61-y female with history of non-Hodgkin Lymphoma and hypothyroidism presenting with fatigue, generalized weakness, nausea, and poor appetite. On admission, physical examination was unremarkable, and labs were notable for hyperkalemia, hyperuricemia and worsening renal function. Imaging showed lymphadenopathy in neck, mediastinum, and left axilla along with mediastinal and retroperitoneal lymphadenopathy. Initial work up of serum electrophoresis was suggestive of multiple myeloma however bone scan did not reveal any lytic lesions. As biopsy results were pending, patient developed worsening cytopenia’s and a repeat serum electrophoresis showed a M spike with IgG lambda against a polyclonal background. Bone marrow biopsy showed polytypic plasmacytosis that was negative for HHV 8 and lymph node biopsy also showed polytypic plasmacytosis. Further work confirmed the diagnosis of idiopathic multicentric Castleman disease and patient was treated with silutuximab and responded well.

Factitious Total Triiodothyronine(T3) Elevation in a Euthyroid Patient With Paraproteinemia

Background: We report a rare case of a patient treated with levothyroxine for hypothyroidism who also had paraproteinemia and was found to have a clinically inconsistent elevation of T3 by RIA. Clinical Case: A 72 Year old African American female with a history of hypothyroidism and IgG kappa Multiple Myeloma (MM) was admitted to the hospital for altered mental status. Her hypothyroidism had been well controlled for years on a stable dose of levothyroxine 0.075mg daily. Review of systems were negative pertinent to thyroid dysfunction. Family history: negative for thyroid disease. On physical exam, an elderly female clinically euthyroid without palpable thyromegaly but was confused and disoriented. Initial Vital signs: BP 117/67, HR 62, RR 14, T 98, SpO2 99% on room air. BMI 19. EKG: normal sinus rhythm, CXR: normal, TSH: 6.27 (0.55-4.78 uIU/mL), total T3: >600 (60-181ng/dL), total T4: 4.5 (3.2-12.6 ug/dL), FT4: 1.22 (0.89-1.76 ng/dL), hemoglobin: 6.6 (12-16 g/dL), hematocrit: 20.9 (38-47 %), Na: 132 (136-145 mmol/L), K: 4.0 (3.5-5.1 mmol/L), Cl: 104 (98-10/ mmol/L), BUN: 45 (20-31 mmol/L), creatinine: 2.8 (0.6-1.0 mg/dL), total protein: 10.1 (5.7-8.2 g/dL), albumin: 1.8 (3.4- 5.0 g/dL), A:G ratio: 0.22 (0.60-1.50 mmol/L). Serum protein electrophoresis revealed Gamma Globulin 5.8 (0.8-1.7 g/dL), Kappa 9270 mg/L (3.3 -19.4 mg/L) Lambda 9.3 (5.7-26.3mg/L), K/L 996.7 (0.26 -1.65), B2 - macroglobulin: 17.9 (ref: <or= 2.51mg/L) which is consistent with an M- spike migrating in the gamma globulin region. Serum TPO, TG, TSI antibodies were negative. Further testing again reported Total T3 >600 with normal reverse T3 21(8-25 ng/dL). Conclusion: Rare cases of factitious elevations of thyroid hormone have been reported in patients with elevated abnormal IgG or IgA proteins having high binding affinity for thyroid hormone.[1] This hypothyroid patient was clinically and biochemically euthyroid except for a dramatic but clinically inconsistent elevated total T3. She also had multiple myeloma with paraproteinemia (IgG Kappa M spike). The few cases reported to date have shown factitious elevation of either or both total T4 and total T3. In our case the factitious elevation was limited to total T3. We alert clinicians to be aware of factitious elevation of thyroid hormones due to high affinity binding to immunoglobulins. In our case this caused spurious elevation of total T3, but not total T4, in a patient with multiple myeloma and an IgG kappa M spike paraprotein. Reference: 1: Marianna Antonopoulou, Arnold Silverberg, “Spurious T3 Thyrotoxicosis Unmasking Multiple Myeloma”, Case Reports in Endocrinology, vol. 2013, Article ID 739302, 3 pages, 2013. https://doi.org/10.1155/2013/739302

Multiple Myeloma with Suspected Non-Secretory Type

Multiple Myeloma (MM) is a hematological malignancy characterized by clonal plasma cell in bone marrow that produceabnormal globulin, which resulted in monoclonal gammopathy. Multiple Myeloma Non-Secretory (MMNS) is a very rareform of multiple myeloma with monoclonal plasmocytic proliferation in bone marrow supported by clinical manifestationand radiological findings. However, plasma cells fail to secrete immunoglobulin. A 44-year-old female came to SardjitoGeneral Hospital with main complaints of weakness and back pain. General weakness and pale palpebral conjunctiva were6 observed (+/+), liver and spleen were not palpable. Blood test results were as follows: Hb 3.0 g/dL, RBC 1.07 x 10 / μL, WBC3 3 562 x 10 /μL, PLT 114 x 10 /μL, A/G ratio 1.07, BUN 51.5 mg/dL, creatinine 4.62 mg/dL, and calcium 3.1 mmol/L. Skeletalsurvey suggested a multiple osteolytic. Protein electrophoresis revealed hypogammaglobulinemia with no M-spike. Therewere 66% of plasma cells in bone marrow. Patient was diagnosed by MMNS. Diagnosis MMNS can be established if clonalplasmacytes is accompanied with renal insufficiency and hypercalcemia. However, monoclonal gammopathy was not foundin serum protein electrophoresis. A case reported of 44-year-old female diagnosed as MMNS with 'punched out' multipleosteolytic, increased plasma cells in bone marrow without evidence of paraprotein in circulation proved by low A/G ratio andnegative M-spike.

Development of a Machine Learning Algorithm for Rapid, Point-of-Care Prediction of Serum Monoclonal Proteins in Multiple Myeloma

Multiple Myeloma (MM) is a cancer of terminally-differentiated plasma cells residing in the bone marrow. Myeloma cells frequently secrete monoclonal proteins that can be used to assess tumor volume and patient response to therapy. Monoclonal proteins are measured by gel electrophoresis and subsequent immunofixation of the observed M-spike for protein typing. However, this a time-consuming process that may take up to 3-5 days that delays physician-patient decision-making, determining response to treatment and can be a significant psychological stressor for patients. Hence, there is an unmet need to develop a more rapid, point-of-care method to determine M-spike levels. Gamma gap is the difference between total serum protein and albumin and includes a variety metabolic proteins, i.e., transferrin, as well as immunologic proteins, e.g., non-involved immunoglobulins, in addition to the M-spike. Since estimation of the non-M-spike portion of the gamma gap cannot be achieved on routine patient care, the gamma gap cannot serve as an accurate surrogate for M-spike protein levels. Here, we hypothesized that an artificial intelligence (AI) algorithm utilizing readily available clinical and laboratory data along with previous and same-day lab variables can accurately predict M-spike levels without the need for serum electrophoresis. Methods: A total of 171 MM patients with 1,472 observations were included in the study, where the upper limit of the observed M-spike was 3.5 gr/dL. Correlation of the observed M-spike with gamma gap was assessed by two correlation methods using the Pearson and Spearman tests. Forty three clinical and lab variables (including total serum protein and albumin) as predictors of M-spike were fed into the machine learning model. Two lagged variables as the last two preceding M-spike values by the same subject were included. When needed, imputation for missing values was applied through interpolation from subject-level linear trend analysis. The random forest model was used, where regression forests are an ensemble of different regression trees and are used for nonlinear multiple regression. The default number of trees was set to be n = 500, and the number of variables considered at each split after random selection was 13. The goal of using a large number of trees was to train enough that each feature had a chance to appear in several models. The data was randomly split into a training set (80%) and a test set (20%), and a regression tree was built with the training set and then validated using the test set. Bootstrapping was used to generate a collection of data sets (n=500), leading to a random forest of regression trees. Results and estimates were combined across trees. Importance was measured by leaving a covariate out of models, and comparing performance with its inclusion. All analyses were performed using R v3.6.2 and its libraries. Results: Median age of the study cohort was 73 years old, range: 42-96), and 44% were male. The median M-spike value was (0.7 gr/dL, range: 0.1-3.5). Fig. 1 shows the number of observations and magnitude distribution for M-spike levels among the patients included in our study. The correlation of the calculated gamma gap and observed M-spike levels was assessed by two methods (Fig.2). The Pearson coefficient was 0.43 for M-spike levels <1 and 0.72 for M-spike levels >1 gr/dL, respectively (Fig.2a). The Spearman coefficient was 0.41 for M-spike levels <1 and 0.74 for M-spike levels >1 suggesting a low overall correlation overall, especially for M-spike levels <1 gr/dL (Fig .2b). In contrast, as shown in Fig. 3, M-spike levels predicted by the AI algorithm (i.e., fitted M-spike in the test set) correlated highly with the observed M-spike levels in the test set (R-square: 94% and RMSE of 0.21). The Pearson and Spearman coefficients were 0.97 and 0.95, respectively. Fig. 3b. Indicates the residual distribution for the RF model with most of values are close to and on both side zero value. Conclusion: Here, we showed that the difference between total protein and albumin (i.e., gamma gap) is a rough estimate of M-spike, especially with lower values. AI algorithm trained by 43 readily available clinical and laboratory variables could predict the observed M-spike very robustly. Taken together, our results indicate that the AI-based method developed here can be further advanced for rapid, accurate, point-of-care measurement of M-spike protein levels in MM patients. Figure 1 Disclosures Malek: Cumberland: Research Funding; Sanofi: Other: Advisory board; Clegene: Other: Advisory board , Speakers Bureau; Takeda: Other: Advisory board , Speakers Bureau; Janssen: Other: Advisory board, Speakers Bureau; Bluespark: Research Funding; Amgen: Honoraria; Medpacto: Research Funding. Caimi:Amgen: Other: Advisory Board; Verastem: Other: Advisory Board; Celgene: Speakers Bureau; Bayer: Other: Advisory Board; ADC Therapeutics: Other: Advisory Board, Research Funding; Kite Pharma: Other: Advisory Board. de Lima:Celgene: Research Funding; Pfizer: Other: Personal fees, advisory board, Research Funding; Kadmon: Other: Personal Fees, Advisory board; Incyte: Other: Personal Fees, advisory board; BMS: Other: Personal Fees, advisory board.