Kalaycio, Matt

Abstract Abstract 532 Primary Plasma cell leukemia (PCL) is a rare plasma cell disorder characterized by an aggressive clinical course. The efficacy of autologous and allogeneic HCT in PCL is uncertain due to the small number of patients that have been reported in the literature. To determine the potential for long-term disease control using HCT in PCL, we performed a retrospective review of 160 patients with primary PCL who received autologous (n=107), allogeneic (n=52) or syngeneic (n=1) HCT and were reported to the CIBMTR between 1995 and 2006. Kaplan–Meier analysis was used for overall survival (OS) and progression free survival (PFS). Cumulative incidence estimates of non-relapse mortality (NRM), relapse, acute and chronic GVHD were calculated. Median age at the time of autologous HCT was 57 years (range, 32-74 years) and allogeneic HCT was 48 years (range, 24-62 years). Median time from initial treatment to autologous HCT was 7 months and allogeneic HCT was 6 months. At the time of transplant, in the autologous HCT group, 19% patients were in complete remission, 56% in partial remission and 1% had relapsed/progressive disease; in the allogeneic HCT group, 17% were in complete remission, 46% in partial remission and 10% had relapsed/progressive disease. In the allogeneic HCT group, 83% of the patients received myeloablative conditioning regimens and 17% received nonmyeloablative/reduced-intensity conditioning. The rate of acute GVHD was 35% (95% confidence interval [CI], 22%-49%) and the rate of chronic GVHD at 3 years was 27% (95% CI, 13%-40%). After a median follow-up of 38 months, 20 of the 52 patients (38%) are alive in the allogeneic HCT group and 68 of the 107 pts (64%) are alive in the autologous HCT group (Figure 1). Progressive disease accounted for 38% of the deaths in the allogeneic HCT group and 86% of the deaths in the autologous HCT group. The outcomes are summarized in table 1. The remarkable overall survival of patients after autologous HCT in this analysis is noteworthy and indicates that autologous HCT is a safe and acceptable treatment option for patients with primary PCL. Although relapse rates are lower, allogeneic HCT carries a much higher risk of NRM and should be considered primarily in the context of a clinical trial. Disclosures: Lonial: Celgene: Consultancy; Millennium: Consultancy, Research Funding; BMS: Consultancy; Novartis: Consultancy; Gloucester: Research Funding.

Abstract Introduction Population level data about the survival outcomes of patients with hematologic malignancies occurring as second cancers is limited. We undertook one of the largest analyses of second hematologic malignancies (SHM) using the Surveillance, Epidemiology & End Results (SEER) registry to identify factors associated with survival. Methods SEER 9 &13 registries were queried to identify adult patients (pts>18 years) diagnosed with SHM that comprised of Acute Myeloid Leukemia (AML), Acute Lymphoid Leukemia (ALL), Myelodysplastic Syndromes (MDS), Chronic Myeloid Leukemia (CML) & Other Acute Leukemias (OAL, biphenotypic leukemias, myeloid sarcomas & leukemias NOS) > 1 year (yr) after diagnosis of a primary unrelated cancer, using ICD-O-3 codes over a 40 yr period (1973-2012). Data on demographics, follow up, primary tumor & survival were abstracted. Univariate & multivariate analyses were performed to identify factors influencing survival of SHM & compared with hematologic malignancies occurring as first cancers. Hazard ratios (HR) with 95% confidence intervals (CIs) are reported & significance was set at p< 0.01. Results SHM cases included 4665 AML, 3569 MDS, 1140 OAL, 1076 CML & 536 ALL pts. Median age at diagnosis of SHM was 73 yrs (range, 18-100) for AML, 79 yrs (18-100) for MDS, 74 yrs(20-99) for OAL, 74 yrs(18-98) for CML & 70 yrs(19-97) for ALL. Pts>60 yrs comprised 79.2% of AML, 92% of MDS, 88.7% of OAL, 82.9% of CML & 75.2% of ALL cases. Gender distribution was similar across all categories, with females comprising 48.5% of AML, 42.2% of MDS, 45.3% of OAL, 42.8% of CML & 50.2% of ALL cases. More than 85% of pts in each group were Caucasians. While the majority of pts had one primary unrelated cancer (82-89%), 14% of AML pts, 17.8% of MDS pts, 13.8% of OAL, 15.3% of CML pts& 10.6% of ALL pts had >1 primary unrelated cancer. Most common primary sites were prostate (19.1-29.6%), breast (16.1-21.9%) & colorectal (9.3-14.5%). Median survival (range) was as follows: AML 0.2 yrs (0-28.2); MDS 1.5 yrs (0-14); OAL 0.2 yrs (0-29.1); CML 2.5 yrs (0-27.1); & ALL 0.7 yrs (0-28.8). Median latency periods (range) for secondary cancers were as follows: AML 5.2 yrs (1.0-24.2); MDS 7.6 yrs (1.0-35.5); OAL 5.7 yrs (1.0-=34.3); CML 5.2 yrs (1.0-31.6); & ALL 6.2 yrs (1.0-35.5). In univariable analyses, age <60 yrs predicted for better survival (p<0.0001) for all SHM, as did diagnosis in recent yrs (2001-2012) (p< 0.0001 for AML, OAL, CML & ALL). Effect of time period on survival for MDS was not done, as it became formally reportable in 2001. Having >2 primaries was associated with an unfavorable outcome for subsequent AML (HR=1.15; 1.08-1.23; p < 0.0001), MDS (HR=1.13; 1.05-1.22: p=0.001) & CML (HR=1.30; 1.13-1.48: p=0.0002). In multivariable analyses, advancing age was an independent adverse prognostic factor for all secondary leukemias & MDS, both as a continuous variable & when categorized as <60 yrs vs. >60 yrs (p<0.001). Race & gender did not impact survival. Using breast cancer as a referent primary, AML pts had worse outcomes following colorectal (HR=1.31; 1.14-1.5; p=0.0002) or lung (HR=1.46; 1.22-1.74; p< 0.0001) primaries, while MDS pts fared worse with prior lung (HR=1.61; 1.22-2.12; p=0.0007) or Non-Hodgkin lymphoma (HR=1.58; 1.3-1.92: p< 0.0001). Of all 2ndCML cases, those after colorectal cancer had worst survival (HR=1.54; 1.12-2.13; p=0.008). Among SHM pts, MDS pts had better survival than those with AML (HR=0.42; 0.4-0.44: p< 0.0001), as did pts with CML (HR=0.37; 0.35-0.41: p< 0.0001) & ALL (HR=0.64; 0.58-0.70; p< 0.0001). Out OAL had similar outcomes compared to AML (HR=0.99; 0.92-1.05; p=0.66). Pts with CML fared better than MDS (HR=0.47; 0.41-0.54; p< 0.0001) & OAL (HR=0.37; 0.34-0.41: p< 0.0001). Pts with ALL had worse survival than CML (HR=1.8; 1.6-2.03: p< 0.0001) & MDS (HR=1.28; 1.11-1.48: p=0.0005), but better than OAL (HR=0.65; 0.58-0.73: p< 0.0001). (Figure 1) Patients with all SHM, except CML, had worse outcomes compared to those with primary hematologic cancers (Figure 2). Conclusion The prolonged latencies for SHM in this study suggest that they are likely therapy-related. Our findings show that type & number of primary cancers influence survival of subsequent leukemias & MDS. SHM have worse survival compared to those with primary hematologic malignancies except CML where non-separation of survival curves suggest that CML as SHM does not behave like a therapy-related neoplasm. Disclosures Sekeres: TetraLogic: Membership on an entity's Board of Directors or advisory committees; Amgen: Membership on an entity's Board of Directors or advisory committees; Celgene Corporation: Membership on an entity's Board of Directors or advisory committees.

Abstract Background: The incidence of thyroid cancers has been rising in the United States primarily due to the increased detection of well differentiated thyroid cancers (WDTC). Radioactive iodine (RAI), which is frequently used to treat WDTC, can be considered a circulating radiation emitter with the potential for mutagenic effects on hematopoietic stem cells. With a growing number of WDTC patients (pts) surviving long term following RAI therapy, establishing the risk of developing a myelodysplastic syndrome (MDS) in this cohort has important clinical implications. Methods: A novel R program, SEERaBomb (Radiat Environ Biophys. 2014;53(1):55-63) was used to query all 18 Surveillance Epidemiology and End Results (SEER) registries to identify WDTC cases treated with RAI and MDS as second cancers. SEERaBomb allows merging of all 18 SEER registries - a function not permitted by the SEER*Stat MP-SIR that limits researchers to use to one of two datasets (SEER 9 or SEER 13), thereby significantly increasing the capture of MDS cases (Figure 1). WDTC includes papillary and follicular thyroid histologies. WDTC survivor person-years at risk for developing MDS was based on age at diagnosis of WDTC, survival time, and age at diagnosis of MDS. Relative risks (RR) for developing MDS were then based on observed/expected cases between the cohorts receiving surgery alone (control), surgery combined with RAI as well as those receiving external beam radiation therapy (EBRT) and surgery with or without (+/-) RAI. Mean radiation exposure to bone marrow was compared between RAI and prostate cancer (as example) treated with other radiation modalities - brachytherapy (seeds), EBRT (3D, IMRT and VMRT). Results: In total 132,157 WDTC patients were identified from 1973-2011, of whom 69,975 (53%) received surgery alone, 59,015 (45%) underwent surgery + RAI and 3167 (2%) underwent EBRT [combined with surgery+/- adjuvant RAI]. Mean age at the time of WDTC diagnosis was: 49.4 years (yrs, range, 2-105) in the surgery group; 46.7 yrs (2-99) in the surgery + RAI group and 55.1 yrs (10-101) in the EBRT group. Median person years of follow up after WDTC diagnosis was: 6.1 yrs (2.5-11.6) in pts undergoing thyroidectomy; 5.4 yrs (2.4-9.7) in the surgery + RAI group and 5.9 yrs (1.7-11) in the EBRT group. A total of 55 patients developed MDS; 24 had surgery alone, 27 received surgery + RAI and 4 had EBRT. Median time to development of MDS after WDTM diagnosis (25-75 interquartile range) was: 6.4 yrs (4-15) in pts receiving surgery alone; 4 yrs (1.7-6.7) in the surgery + RAI group and 3.6 yrs (2.4-7.9) in the EBRT group. MDS cases were classified as RA (n=11), RARS (n=2), RCMD (n=3), RAEB (n=9), MDS with 5q deletion syndrome (n=1), and MDS-U (n=29). Compared to patients definitively treated with surgery, those who additionally received RAI (+/- EBRT) had a statistically significantly increased risk of developing MDS within the first two years of exposure (RR=1.9; 95% CI, 0.43-5.61 vs. RR=5.8; 95% CI, 2.84-12.22). Beyond 2 years, the risk for MDS drops to baseline rates, and a trend observed beyond 12 years, which did not reach statistical significance (Figure 2). Conclusion: WDTC pts undergoing RAI treatment appear to have an increased risk of developing MDS within the first two years of exposure to RAI. It is difficult to determine whether this early excess risk is true risk or an effect of ascertainment bias (RAI pts followed more closely), as the kinetics of RAI indicate a lower magnitude of bone marrow exposure compared to other radiation modalities (Figure 3). Considering the long latency of MDS seen in atomic bomb cohorts, relative young age of WDTC pts, current trend towards overdiagnosis and overtreatment of WDTCÕs, MDS rates are likely to continue to rise. Disclosures Sekeres: Celgene Corporation: Membership on an entity's Board of Directors or advisory committees; Amgen: Membership on an entity's Board of Directors or advisory committees; TetraLogic: Membership on an entity's Board of Directors or advisory committees.

Abstract Introduction: Surveillance, Epidemiology, and End Results (SEER) registry captures data on acute leukemias (AL) that morphologically do not conform to well characterized acute leukemic entities in the established French-American-British (FAB) nomenclature. Not much is known about population level outcomes of these rare AL. We performed a comprehensive survival analysis of these rare AL over a forty-year period using the SEER registry. Methods: SEER 18 dataset was queried to identify adult patients (pts > 18 years) diagnosed with Acute Myeloid Leukemia (AML), Acute Lymphoblastic Leukemia (ALL) and rare AL from 1973-2012. Rare AL cases were identified using ICD-O-3 morphology codes and included Leukemias, Not Otherwise Specified (NOS) (9800-01, 9860) and myeloid sarcoma (9930). Data on demographics,follow up and survival was abstracted. Univariate and multivariate analysis was performed to identify factors influencing survival of rare AL and survival patterns were compared to AML and ALL. Hazard ratios (HR) with 95% confidence intervals (CIs) are reported. Results: We identified 61,550 AML pts, 13,457 ALL pts, 525 myeloid sarcoma pts and 13,446 pts with leukemia NOS. Median age at diagnosis was 68 years (yrs, range 18-100) for AML, 54 yrs (18-100) for ALL, 65 yrs (18-95) for myeloid sarcoma and 76 yrs (18-100) for leukemia NOS. Gender distribution was similar with females comprising 45.9% of AML, 43% of ALL, 42.3% of myeloid sarcoma and 47.2% of leukemia NOS cases. Over 80% pts in all categories were Caucasian. Median survival was as follows: 0.4 (0-40) yrs for AML; 1.4 (0-39.8) yrs for ALL; 0.7 (0-32.9) yrs for Myeloid Sarcoma; and 0.3 (0-38.9) yrs for leukemia NOS. In univariable analysis, younger age predicted for better survival for both myeloid sarcoma and leukemia NOS (p<0.0001). Improvement in survival over time period was observed for pts with leukemia NOS (p < 0.0001). In multivariable analysis, advancing age was an independent prognostic factor for rare AL, both as a continuous variable and when categorized as older vs. younger pts using age cutoff of 60 yrs. Compared to younger pts, HR for pts > 60 yrs with rare AL was as follows: 2.68 (2.13-3.37; p <0.0001) for myeloid sarcoma; and 2.18 (2.07-2.31; p<0.0001) for leukemia NOS. No difference in survival over time was seen for myeloid sarcoma. Pts with leukemia, NOS diagnosed in yrs 2006-2012 (HR=0.87; 0.82-0.92; p < 0.0001) had better survival compared to the referent years 1973-80. Race and gender did not impact survival. In multivariable analysis, pts with leukemia NOS did worse overall compared to AML (HR=1.13; 1.11-1.53; p<0.0001) and ALL (HR=1.86; 1.81-1.91; p<0.0001). Interestingly, younger pts had a poorer outcome than AML (HR=1.10; 1.04-1.16; p=0.0003), but pts > 60 yrs had a better outcome (HR=0.93; 0.91-0.95; p<0.0001). Both younger (HR=1.33; 1.26-1.41; p<0.0001) and older pts (HR=1.45, 1.40-1.50: p<0.0001) compared unfavorably to ALL (Figure 1). Pts with myeloid sarcoma had a better survival compared to AML pts (HR=0.77; 0.7-0.85; p<0.0001) (Figure 2). This was uniform for pts <60 yrs (HR=0.77; 0.64-0.93; p=0.006) and > 60 yrs (HR= 0.77; 0.69-0.87; p<0.0001). Conclusion: Pts with leukemias NOS have poor outcomes compared to AML and ALL but prognosis has improved in recent years. The most probable explanation is identification of prognostic molecular mutations that has allowed accurate risk-stratification of pts using the WHO classification and in turn better selection of treatment strategies. Several reasons have limited our understanding of these rare AL - small numbers, exclusion from clinical trials and wide variability in treatment options used. Efforts at better molecular characterization of these AL holds the most promise for improving outcomes. Disclosures Sekeres: Celgene Corporation: Membership on an entity's Board of Directors or advisory committees; TetraLogic: Membership on an entity's Board of Directors or advisory committees; Amgen: Membership on an entity's Board of Directors or advisory committees.

Abstract Abstract 5045 Observation remains the standard approach to patients with monoclonal gammopathy of undetermined significance (MGUS) and asymptomatic multiple myeloma (AMM) who lack end-organ impairment characteristic of symptomatic MM. Institution of early MM therapy has not been shown to improve survival or alter the natural progression of the disease. Cyclooxygenase-2 (COX2) inhibitors inhibit proliferation and induce apoptosis in drug resistant MM cell lines. 23 asymptomatic adult patients with a serum monoclonal protein >1g/dL, from Cleveland Clinic, Mayo Clinic, or University of Arkansas Medical Center between August 2005 and April 2008 were randomized to either celecoxib or placebo in a double-blind, 1:1 fashion. The primary end point was change in the monoclonal protein concentration after 6 months. Correlative biomarkers, including IL6 and IL-6R, CD4/CD8+ T-cell ratios, and beta-2 microglobulin, were examined before and after treatment. Celecoxib had no significant effect on the median monoclonal protein concentration, which went from a median of 1. 44 to 1. 65 g/dL with placebo and 2. 42 to 2. 24 g/dL with Celecoxib (p=0. 36), or on any of the analyzed correlative markers (all p-values >0. 34). One patient's monoclonal protein concentration increased on placebo, but he did not require treatment for symptomatic MM. There were no adverse events of Grade 2 or higher. Celecoxib was well tolerated in contrast to previous reports when given in higher doses and in combination with thalidomide. Although accrual was limited we conclude that celecoxib has no clinical or immunologic effect on high risk asymptomatic plasma cell disorders. This study does not support using celecoxib for preventing progression to symptomatic MM. Future studies on progression of MGUS to symptomatic MM would require more patients and a longer follow-up period. Disclosures: Off Label Use: Celecoxib for chemoprophylaxis. Hussein:Celgene: Employment. Reu:Celgene: Research Funding.

Abstract Abstract 3925 Introduction: CLL patients (pts) who relapse following purine-analog or bendamustine-based treatments have a poor prognosis. These pts have limited therapeutic options and novel agents with alternative mechanisms of action are needed. Several phase 1 and 2 trials in rel/ref CLL showed promising activity with escalating dose regimens of lenalidomide (LEN). In other phase II studies improved clinical responses to lenalidomide appeared to correlate with dose levels > 5mg/day. This phase 2 trial investigates the safety of LEN initiated at 3 different starting doses followed by a step-wise dose escalation as tolerated in rel/ref CLL. Methods: In this ongoing trial, eligible pts with rel/ref CLL who have received ≥ 1 prior treatment regimen containing purine-analog or bendamustine are being enrolled. The objectives of this study are to evaluate primarily the safety and secondarily the efficacy of different LEN dose regimens. Pts are randomized 1:1:1 to receive a double-blinded starting dose of 5 mg, 10 mg, or 15 mg oral LEN on days 1–28 of each 28-day cycle. In all 3 treatment arms, the dose is escalated by 5 mg increments every 28 days to reach a maximum dose of 25 mg/d, depending on tolerability. In instances of poor tolerability, dose reductions also occur in 5 mg steps. Pts are stratified by relapsed versus refractory status to their last purine-analog or bendamustine-based treatment regimen and according to age (< 65 vs ≥ 65 years). Tumor lysis syndrome (TLS) prophylaxis comprises of oral hydration and allopurinol 300 mg/day and is initiated ≥ 3 days prior to starting study drug and for a minimum of the first 3 treatment cycles. A total of 105 pts are planned for enrollment to the study. Per protocol, unblinded interim analyses were conducted by the independent Data Monitoring Committee (DMC) after 18 subjects completed 1 cycle and continue at 13-week intervals. Results: To date, 95 pts are enrolled at a median age of 64 years (range 32–81). Enrolled pts are primarily male (67%) and Caucasian (92%). Cytogenetic data are available for 73 pts; 21% have del(17p), 55% have del(13q), 25% have del(11q), and of 72 patients evaluable for trisomy 12, 11 patients (15%) tested positive. IGVH is unmutated in 77% of 77 evaluable pts and 44% of 84 evaluable pts are ZAP70-positive. Based on the Binet and Rai staging systems 9 (10%), 25 (26%) and 24 (25%) of subjects are stage A, B and C, respectively; 7 (7%), 12 (13%) and 16 (17%) subjects are low, intermediate or high-risk disease, respectively. For 2 (2%) subjects the Binet/Rai staging is currently unknown. Overall, 19 pts (20%) received prior bendamustine-containing treatment and 71 pts (75%) received prior fludarabine-based treatment. The median number of prior therapies was 3 (range 1–10). Most common hematological grade ≥ 3 AEs include neutropenia (62%) and thrombocytopenia (34%). At baseline, 19% of pts presented with grade 1–2 neutropenia. The most common non-hematological ≥ grade 3 AEs include pneumonia (13%), tumor flare (13%), and fatigue (11%). TLS was reported in 3 pts (3%): grade 1, 3, and 4. In total, 8 grade 5 events were reported, 3 of which were suspected to be related to LEN: 2 cases of pneumonia and 1 death for unknown cause. At the time of the cut-off, 59 pts (62%) have discontinued treatment. Most common reasons for treatment discontinuation include disease progression (n = 20) and AEs (n = 20). To date, 47 pts (49%) have dose escalated above their starting dose levels of which 12 patients escalated to the highest dose level (25 mg daily). 18 subjects have had no dose level reduction or escalations and 1 patient is still in the first cycle of the study. Average duration of treatment is 6.5 cycles, and median number of cycles is 4. Efficacy evaluations are completed monthly after 3 months of study drug treatment. At time of the data reporting, 5 pts were on study drug but did not reach the first assessment at cycle. For the 90 pts evaluable for response, the investigator's assessment indicates 2 pts (2%) reached CR, 36 (40%) achieved PR, 33 (37%) patients had SD, and 19 (21%) pts progressed. Conclusion: The independent DMC, as of 14 June, 2012 (N=95), recommended that accrual into all three treatment arms should continue as planned, suggesting all three starting doses were well tolerated. To date, the ORR is 42% and 49% of pts were dose escalated at least once. In this rel/ref CLL population LEN appears active, and completion of accrual will clarify the appropriate dose at which to initiate therapy. Disclosures: Wendtner: Celgene Corporation: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Roche: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; GSK: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Mundipharma: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. Off Label Use: use of lenalidomie in relapsed/refractory CLL. Hallek:Celgene Corporation: Consultancy, Honoraria, Research Funding. Hillmen:Celgene Corporation: Honoraria. Gregory:Celgene Corporation: Honoraria, Research Funding. Stilgenbauer:Celgene Corporation: Honoraria, Research Funding, Speakers Bureau. Kipps:Celgene Corporation: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. Purse:Celgene Corporation: Employment, Equity Ownership. Zhang:Celgene Corporation: Employment, Equity Ownership. Mei:Celgene Corporation: Employment.

Abstract Allogeneic hematopoietic cell transplantation (allo-HCT) is a potentially curative treatment of chronic myeloid leukemia (CML). Optimal conditioning intensity for allo-HCT for CML in the era of tyrosine kinase inhibitors (TKIs) is unknown. Using the Center for International Blood and Marrow Transplant Research database, we sought to determine whether reduced-intensity/nonmyeloablative conditioning (RIC) allo-HCT and myeloablative conditioning (MAC) result in similar outcomes in CML patients. We evaluated 1395 CML allo-HCT recipients between the ages of 18 and 60 years. The disease status at transplant was divided into the following categories: chronic phase 1, chronic phase 2 or greater, and accelerated phase. Patients in blast phase at transplant and alternative donor transplants were excluded. The primary outcome was overall survival (OS) after allo-HCT. MAC (n = 1204) and RIC allo-HCT recipients (n = 191) from 2007 to 2014 were included. Patient, disease, and transplantation characteristics were similar, with a few exceptions. Multivariable analysis showed no significant difference in OS between MAC and RIC groups. In addition, leukemia-free survival and nonrelapse mortality did not differ significantly between the 2 groups. Compared with MAC, the RIC group had a higher risk of early relapse after allo-HCT (hazard ratio [HR], 1.85; P = .001). The cumulative incidence of chronic graft-versus-host disease (cGVHD) was lower with RIC than with MAC (HR, 0.77; P = .02). RIC provides similar survival and lower cGVHD compared with MAC and therefore may be a reasonable alternative to MAC for CML patients in the TKI era.

Purpose To investigate the risk and outcomes of second hematologic malignancies (SHMs) in a population-based cohort of patients with well-differentiated thyroid cancer (WDTC) treated or not with radioactive iodine (RAI). Methods Patients with WDTC were identified from SEER registries. Competing risk regression analysis was performed to calculate the risks of SHMs that occurred after WDTC treatment and outcomes after SHM development were assessed. Results Of 148,215 patients with WDTC, 53% received surgery alone and 47% received RAI. In total, 783 patients developed an SHM after a median interval of 6.5 years (interquartile range, 3.3 to 11.2 years) from WDTC diagnosis. In multivariable analysis, compared with those undergoing thyroidectomy alone, RAI treatment was associated with an increased early risk of developing acute myeloid leukemia (AML; hazard ratio, 1.79; 95% CI, 1.13 to 2.82; P = .01) and chronic myeloid leukemia (CML; hazard ratio, 3.44; 95% CI, 1.87 to 6.36; P < .001). This increased risk of AML and CML after RAI treatment was seen even in low-risk and intermediate-risk WDTC tumors. Occurrence of AML but not CML in patients with WDTC was associated with shorter median overall survival compared with matched controls (8.0 years v 31.0 years; P = .001). In addition, AML developing after RAI trended toward inferior survival compared with matched controls with de novo AML (median overall survival, 1.2 years v 2.9 years; P = .06). Conclusion Patients with WDTC treated with RAI had an increased early risk of developing AML and CML but no other hematologic malignancies. AML that arises after RAI treatment has a poor prognosis. RAI use in patients with WDTC should be limited to patients with high-risk disease features, and patients with WDTC treated with adjuvant RAI should be monitored for myeloid malignancies as part of cancer surveillance.