Gertz, Morie A.

e20520 Background: Cytogenetic abnormalities detected by FISH are found in the majority of multiple myeloma (MM) patients. Although their prognostic value has been studied extensively, less is known about their association with disease characteristics and treatment response. Methods: To address these questions, we designed a retrospective study including 2031 Mayo Clinic patients diagnosed with MM from 2004 to 2018. We compared baseline characteristics and treatment outcomes between primary cytogenetic groups: t(11;14), t(4;14), t(14;16), (14;20), t(6;14), unknown IgH translocation/del and trisomy (without IgH translocation). These included 373, 177, 78, 20, 18, 228 and 791 patients respectively. Kruskal-Wallis and Fisher’s exact tests were used for categorical and continuous variables respectively. Time to next treatment (TTNT) was estimated using Kaplan-Meier method and compared using Log-Rank test. Results: t(4;14), t(14;16), t(6;14) and t(14;20) groups were associated with hemoglobin < 10 g/dL, beta2microglobulin > 5.5 µg/ml, ISS stage 3 and ≥50% bone marrow plasma cells. The latter 3 groups were also associated with renal dysfunction (Cr ≥2 mg/dL) and higher urinary monoclonal protein. t(4;14) was associated with IgA isotype, serum monoclonal protein ≥1g/dL and plasma cell labeling index ≥1%. Light chain myeloma was more prevalent in patients with t(11;14). Overall response rate (ORR) to proteasome inhibitor (PI) induction was higher for those with IgH translocations (any) compared to trisomies (85% vs 77% P = 0.02), while ORR was higher for those with trisomies with immunomodulatory drug (IMiD) induction (90% vs 78% P < 0.01). The rate of ≥ very good partial response was higher for patients with high risk IgH translocations [t(4;14), t(14;16) or t(14;20)] compared to standard risk with PI-IMiD combination treatment (88% vs 65% P < 0.01). Otherwise, response rates did not differ between these 2 groups. TTNT was longer in patients with trisomies compared to those with IgH translocation with IMiD or PI-IMiD treatments (32.2 vs 19 and 44 vs 27.4 months, respectively P < 0.01). For all cytogenetic groups, better treatment responses and longer TTNT were seen with PI-IMiD combinations compared to other treatments. Conclusions: t(4;14), t(14;16), t(6;14), and t(14;20) are associated with high risk disease characteristics. Patients with IgH translocations may have better response to PI induction compared to those with trisomies, while those with trisomies may have better response to IMiD treatment, with best outcomes for both seen with PI-IMiD combinations.

e20515 Background: IgA monoclonal proteins (MCPs), unlike IgG MCPs, often migrate in the beta region on serum protein electrophoresis (SPEP) which can lead to underestimation of their size due to the co-migration with physiologic proteins. In IgA multiple myeloma (MM), the utility of quantitative IgA levels in assessing disease response in comparison to SPEP is not well studied. Methods: We retrospectively analyzed 304 IgA MM patients, diagnosed between 2004 and 2018, with available serial MCP and quantitative IgA levels. Kaplan Meier analysis was used to estimate the median progression free survival (mPFS) using the IMWG criteria and our study definition of IgA progression (2 consecutive IgA values that are > 25% above the nadir IgA value and above upper limit of normal (ULN) of 356 mg/dL, and a detectable IgA MCP on serum immunofixation). The mPFS was defined as the time from treatment initiation until disease progression or death. Results: IgA MCP migrated in the beta region in 134 (44%) patients, and in the gamma region in 150 (56%) patients. At diagnosis the median MCP was 3 (IQR 1.9-4) g/dL and the median IgA was 3240 (IQR 2008-4420) mg/dL. The median time from treatment initiation to MCP nadir was 80 (IQR 42-144) days and median time to IgA nadir was 154 (IQR 90-238) days. At MCP nadir 40% of patients had an IgA above the ULN. All complete responders (n = 104) had normal IgA levels, with a median IgA of 54 (IQR 27-88) g/dL. A ≥90% decrease in IgA between treatment initiation and IgA nadir, compared to a < 90% decrease, was associated with a longer mPFS (34 vs. 20 months, p = 0.006) and overall survival (97 vs. 33 months, p = 0.003). Patients with serial MCP and IgA levels available prior to progression (n = 195) were used to compare the mPFS using the IMWG and IgA progression criteria. The mPFS using the IgA criteria was 32 (95% CI 29-39) months, versus 39 (95% CI 33-45) months using IMWG criteria. Overall, 92 (47%) patients progressed by both IMWG and IgA criteria. At the time of progression using the IgA criteria compared to at IMWG progression, the median hemoglobin was higher (13.3 vs. 11.6 g/dL, respectively, p < 0.001) and fewer patients had new symptomatic bone lesions (2% vs. 33%, respectively, p < 0.001). Conclusions: In IgA MM patients, monitoring quantitative IgA levels predicts disease response and allows for earlier detection of disease progression, prior to the development of end organ damage.

e20510 Background: pPCL is a rare and aggressive form of multiple myeloma (MM) with dismal survival outcomes compared to the remainder of MM patients. Several studies have validated the optimal cutoff for defining pPCL to be >5% circulating plasma cells (cPCs) on a peripheral blood smear due to equally poor outcomes in this less restrictive cohort. We evaluated the clinical outcomes and cytogenetic features of patients diagnosed with pPCL at our institution that were treated with novel agent induction therapies. Methods: We evaluated patients with pPCL diagnosed between 2000 – 2018 (using the 5% cut off) at Mayo Clinic, Minnesota. Data was extracted from a prospectively maintained database and from the review of electronic medical records. Patients were categorized as having high risk (HR) cytogenetics if any of the following abnormalities were present: del 17p, t(4;14), t(14;16) or t(14;20). Chi-square tests and Fisher exact tests were used to compare differences between sub-groups. Survival analysis was performed by the Kaplan-Meier method and differences assessed using the log rank test. Results: This cohort consisted of 67 patients with pPCL with a median age of 62 years (range: 34-91) of which 33 (46%) were male. The median follow up was 46 months (95% CI: 41 – 90). The median bone marrow plasma cell involvement was 84% (Range: 10 – 100) and the median cPCs percentage on the peripheral blood was 23% (range: 5 - 93). Data on primary cytogenetic abnormalities were available in 60 (85%) patients and the distribution was as follows: t(11;14) – 27 (45%), t(4;14) – 5 (8%), t(14;16) – 8 (13%), t(14;20)- 3(5%) and del 17p- 15(25%). All patients received novel agent induction therapy with 36 (54%) having received autologous stem cell transplant and 4 undergoing an allogeneic stem cell transplant. The median time to next therapy (TTNT) and overall survival (OS) for all patients was 13 months (95% CI: 9 – 17) and 24 months (95% CI: 19 – 40) respectively; 16 months and 51 months for standard risk (SR) vs. 10 months and 19 months for HR (P = 0.005 for OS), when stratified by cytogenetic risk. There were only 17 (31%) patients who were alive for ≥48 months since diagnosis (i.e. twice the median OS of this cohort); The absence of high risk cytogenetics (P = 0.049) and non-elevated LDH level (P = 0.047) at diagnosis predicted for achieving this survival goal. Conclusions: The outcomes of patients diagnosed with pPCL remain poor despite the use of novel agent induction therapy. However, some patients appear to do better than expected and this phenomenon may be influenced by the presence of HR cytogenetics.

e20506 Background: Autologous stem cell transplantation (ASCT) prolongs survival in patients with light chain (AL) amyloidosis. Mayo 2012 stage and increased plasma cell percentage (%PC) are known predictors for survival. Increased beta-2 microglobulin (B2M) predicts survival in patients with multiple myeloma. However, its prognostic effect in patients with AL amyloidosis undergoing ASCT is not known. Methods: We retrospectively reviewed patients who had a diagnosis of AL amyloidosis and were treated with ASCT between July-1996 and September-2017. Patients with creatinine > 1.2 mg/dL were excluded, as that affects B2M levels. The receiver operator curve was used to determine the best cutoff for B2M in predicting survival and was 2.5 mcg/mL. Baseline characteristics were compared between patients with B2M > 2.5 and ≤2.5. Progression-free survival (PFS) was defined as time from ASCT to relapse or death, whichever occurred first. Overall survival (OS) was calculated from ASCT to death of any cause. Univariate and multivariate analysis were done for OS. Results: Five-hundred patients were identified and 222 (44%) had a B2M > 2.5. These patients were more likely to be > 65 years old (32% vs. 17%, P = 0.0001), have Mayo 2012 stage III/IV (33% vs. 8%, P < 0.0001), have ≥3 organs involved (25% vs. 14%, P = 0.001), and have ≥10% PCs (56% vs. 40%, P = 0.0002) compared to patients with B2M ≤2.5. The median PFS and OS were shorter in patients with B2M > 2.5 (median PFS: 64 vs. 80 months, P = 0.03); (median OS: 104.9 vs. 175.5 months, P < 0.0001). On univariate analysis, predictors for OS included age > 65 (HR: 1.6, P = 0.001), Mayo 2012 stage III/IV (HR: 3.3, P < 0.0001), ≥3 organs involved (HR: 1.3, P = 0.06), ≥10% PC (HR: 1.5, P = 0.004), melphalan conditioning 200mg/m2 (HR: 0.28, P < 0.0001), and B2M > 2.5 (HR: 1.8, P < 0.0001). In a multivariate analysis, only Mayo 2012 stage III/IV (HR: 1.8, P = 0.006), melphalan conditioning 200mg/m2 (HR: 0.35, P < 0.0001), and B2M > 2.5 (HR: 1.7, P = 0.01) remained independent predictive of OS. Conclusions: Beta-2 microglobulin > 2.5 is an independent predictor for OS in AL amyloidosis patients undergoing ASCT and should be routinely measured.

8550 Background: Waldenström Macroglobulinemia (WM) is a rare lymphoplasmacytic malignancy characterized by the presence of a recurrent point mutation in the MYD88 gene (MYD88L265P) in 80-95% of cases. Patients with MYD88WT genotype comprise a small subset that responds poorly to ibrutinib and other Bruton tyrosine kinase inhibitors. We examined the characteristics and outcome of WM patients with MYD88WT genotype predominantly treated with non-BTK inhibitor based therapies. Methods: Patients with a diagnosis of WM seen at Mayo Clinic, Rochester, between 1996 and 2018 were included. Their characteristics and outcomes were assessed from the time of active disease. Marrow MYD88 genotyping was assessed with an allele specific PCR assay (analytic sensitivity 1%). Categorical and continuous variables were compared using Chi square and Wilcoxon tests, respectively. Time-to-event analyses were performed using Kaplan Meier test. Results: Of 986 patients with active WM, MYD88 genotype data were available in 331 (34 %) patients; 72 (22%) and 260 (78%) patients harbored MYD88WT and MYD88L265P genotypes, respectively. The median follow-up was 5.8 years (95% CI: 5.0-6.5 years) from active WM; 6 years MYD88WT vs 5.4 years for MYD88L265P cohort. Median age was 63 years and 66 years in the MYD88WT and MYD88L265P cohorts, respectively (p = 0.07) with 46% and 53% patients being ≥65 years of age, respectively (p = 0.36). Pre-treatment marrow lymphoplasmacytic (LPL) infiltrate (median 40% for MYD88WT vs 60% for MYD88L265P; p = 0.001) and beta-2 microglobulin (median 3 µg/mL for MYD88WT vs 3.9 µg/mL for MYD88L265P; p = 0.02) were lower in the MYD88WT compared to the MYD88L265P cohort; other laboratory parameters at active disease were comparable. Per IPSSWM prognostic criteria, MYD88WT had fewer patients in the high risk group (18% for MYD88WT vs 42% MYD88L265P; p = 0.03). Patients with MYD88WT had higher likelihood of histological transformation [18% for MYD88WT vs 4% for MYD88L265P; odds ratio 5.8 (95% CI: 2.5-13.5; p < 0.0001)]. Among patients with treatment data available, only 35 (11%) patients received ibrutinib. The 5-year overall survival (OS) from active disease was comparable (85% in MYD88WT vs 82% in MYD88L265P cohort; p = 0.7). Conclusions: MYD88WT genotype in WM is associated with lower marrow LPL infiltration, lower likelihood of high-risk IPSSWM categorization and a higher likelihood of histological transformation in comparison to MYD88L265P mutant subpopulation. MYD88 genotype does not affect the OS from active disease in predominantly non-BTK inhibitor treated patients.

8549 Background: Proteinuria evaluation is essential for diagnosing and monitoring of renal involvement in light chain (AL) amyloidosis. A 24 hour protein collection (24h UP) is the gold standard for proteinuria assessment however it is cumbersome and can be inaccurate. A spot urine albumin to creatinine ratio (uACR) has been proposed as a convenient method to estimate 24hUP. We aimed to validate the correlation between uACR and 24hUP in a large cohort of patients. Methods: We retrospectively studied systemic AL amyloidosis patients evaluated between 2010 and 2019 at Mayo Clinic, with a uACR and 24hUP collected less than 7 days apart. Linear regression analysis was used to construct a prediction model for 24hUP with uACR as the primary predictor. Possible confounders (age, gender, body mass index, morning versus afternoon spot urine collection, estimated glomerular filtration rate) for the primary relationship between uACR and 24h UP were evaluated in the model. We used receiver operating characteristic (ROC) analysis to identify the best uACR cutoff to predict significant proteinuria (defined as a 24hUP > 500mg). Results: We included 665 patients, with a median age of 66 years (IQR 59-72). The spot urine was collected in the morning (before 1200 hours) in 382 (57%) patients, and in the afternoon in 283 (43%) patients. The median 24hUP was 321 (IQR 129-2512.5) mg, median uACR was 107 (IQR 13.5-1845) mg/g, and median serum creatinine was 1.2 (IQR 1-1.8) mg/dL. The uACR correlated well with 24h UP (Pearson’s r= 0.83, 95% CI 0.80-0.85). Linear regression showed that E (24h UPi) = 362 + 1.05(uACRi), and this model was statistically and clinically significant (p < 0.001 and R2 of 0.68, respectively). Age, gender, body mass index, eGFR, and time of day of spot urine collection did not confound the primary relationship between uACR and 24hUP, and no collinearity was observed. A uACR cutoff of > 280 mg/g was the best predictor of a 24hUP > 500 mg (area under the ROC curve 0.98, sensitivity 92%, specificity 97%). For simplicity, we assessed the predictive value of uACR > 300 mg/g for 24h UP > 500 mg. Among patients with 24huACR > 300 mg/g 264 (96%) had a 24hUP > 500 mg, and 31 (7%) of patients with uACR < 300 mg/g had a 24h UP > 500 mg (p < 0.001). Conclusions: In systemic AL amyloid patients, we showed that uACR on a random urine sample correlated well with 24h UP, and can be used to estimate proteinuria with a linear regression model. Based on these findings, and the convenience of uACR testing for patients, we propose that uACR should be used to monitor renal response to AL amyloidosis therapy.

8516 Background: The role of induction therapy prior to autologous stem cell transplant (ASCT) in immunoglobulin light chain (AL) amyloidosis remains controversial. Data on the prognostic impact of response to induction in a transplanted cohort are lacking. The aim of this study was to assess the impact of response to induction therapy on survival in patients undergoing ASCT for AL amyloidosis. Methods: We conducted a retrospective study of all newly diagnosed AL amyloidosis patients who received induction prior to ASCT between January 2007 and August 2017 at Mayo Clinic, Rochester, Minnesota. Patients receiving only corticosteroids prior to transplant were excluded as were those with an involved light chain of less than 5 mg/dL (not measurable for response). Results: 134 patients met inclusion criteria. The median age at diagnosis was 60 (range 36-74) and 85 (63%) were men. The most commonly used induction regimen was proteasome inhibitor-based (73.1%, n=98). The overall response rate to induction was 83% (complete response 17%, very good partial response 30% and partial response 36%). With a median follow up of 56.5 months, the median PFS and OS was 48.5 months and not reached, respectively. Response depth to induction therapy was associated with improved PFS and OS and was independent of the bone marrow plasma cell percentage. The median PFS was not reached for patients achieving ≥VGPR prior to ASCT and 33.8 months for patient achieving PR or less (P=0.001). The median OS was longer in patients with deeper responses (not reached for patients achieving ≥VGPR vs. 128 months for patients achieving PR or less (P=0.02). On multivariable analysis, independent predictors of OS were melphalan conditioning dose (RR= 0.38; P=0.018) and depth of response prior to transplant (RR 2.52; P=0.039). Conclusions: Hematologic response prior to transplant predicts post-transplant outcomes in patients with AL amyloidosis. [Table: see text]

e19112 Background: We sought to study the longitudinal QOL, financial burden and perception of care (PoC) in patients with multiple myeloma (MM) or lymphoma (LYM) enrolled on CTs vs. those receiving standard care (Non-CT). Methods: QOL was evaluated using the FACT-G questionnaire (baseline, 1, 2, 3, 6 & 12 months). Financial burden was evaluated using 10 questions adapted from the MEPS survey and PoC was evaluated using 6 questions adapted from the CAHPS survey (baseline, 3, 6 & 12 months). Results: 35 (28%) of 123 enrolled patients were treated on CTs. Demographics, QOL and financial burden are shown in the Table. Baseline QOL scores were similar in CT vs non-CT groups. Age, gender and self-reported performance status were significantly associated with baseline QOL, while lines of therapy & cancer type were not. There were no significant differences in change in QOL vs. baseline in patients treated on CTs vs. non-CTs at 1, 2, 3, 6 and 12 months from baseline, including the total FACT-G score and the subdomains of functional, physical, emotional and social well-being. Within each group, there was a decline in QOL over the first 3 months, specifically physical WB, which gradually returned to baseline. Baseline answers to financial burden questions were similar, except patients on CTs reported less need for taking extended time-off from work (22% vs. 46%, p = 0.02). The only statistically significant difference in individual Qs over 1 year was ‘worry about having to pay large medical bills’ at the 6-month timepoint (CT: 0%, non-CT: 52%, p = 0.01). Using a composite score from 4 MEPS questions, financial burden over time was lower in the CT group, though the differences were not statistically significant. There were no significant differences in PoC at baseline, 3, 6 and 12 months in the two groups based on answers to each of the six questions adapted from the CAHPS survey. Conclusions: Over 1 year, patients on CTs experienced a similar QOL and perceived their care to be similar as non-CT patients. Financial burden was lower in the CT group, though differences were not statistically significant. [Table: see text]

AbstractMetaphase cytogenetic abnormalities, plasma cell proliferation index (PCPro), and gain 1q by fluorescence in situ hybridization (FISH) are associated with inferior survival in newly diagnosed multiple myeloma (MM) treated with novel agents; however, their role in risk stratification is unclear in the era of the revised International Staging System (R-ISS). The objective of this study was to determine if these predictors improve risk stratification in newly diagnosed MM when accounting for R-ISS and age. We studied a retrospective cohort of 483 patients with newly diagnosed MM treated with proteasome inhibitors and/or immunomodulators. On multivariable analysis, R-ISS, age, metaphase cytogenetic abnormalities (both in aggregate and for specific abnormalities), PCPro, and FISH gain 1q were associated with inferior progression-free (PFS) and overall survival (OS). We devised a risk scoring system based on hazard ratios from multivariable analyses and assigned patients to low-, intermediate-, and high-risk groups based on their cumulative scores. The addition of metaphase cytogenetic abnormalities, PCPro, and FISH gain 1q to a risk scoring system accounting for R-ISS and age did not improve risk discrimination of Kaplan-Meier estimates for PFS or OS. Moreover, they did not improve prognostic performance when evaluated by Uno’s censoring-adjusted C-statistic. Lastly, we performed a paired analysis of metaphase cytogenetic and interphase FISH abnormalities, which revealed the former to be insensitive for the detection of prognostic chromosomal abnormalities. Ultimately, metaphase cytogenetics lack sensitivity for important chromosomal aberrations and, along with PCPro and FISH gain 1q, do not improve risk stratification in MM when accounting for R-ISS and age.