Neel K. Gupta

Differential effects of neurotoxic destruction of descending noradrenergic pathways on acute and persistent nociceptive processing

Although many pharmacological studies indicate that bulbospinal noradrenergic projections contribute to antinociception, lesions of the major brainstem noradrenergic cell groups have provided conflicting evidence. Here we used a new immunotoxin, anti-dopamine beta-hydroxylase-saporin, to re-examine the contribution of noradrenergic pathways to nociception and to morphine analgesia. We treated rats intrathecally by lumbar puncture with the immunotoxin and examined dopamine beta-hydroxylase (DbetaH) immunoreactivity seven and 14 days after treatment. There was no change in DbetaH staining at 7 days; however, 14 days after treatment we demonstrated significant destruction of noradrenergic neurons in the locus coeruleus and in the A5 and A7 cell groups. There was a concomitant loss of noradrenergic axons in the dorsal and ventral horns of the lumbosacral and cervical cord. Consistent with the lack of anatomical changes, we found no difference in nociceptive responses in the hot-plate, tail-flick or formalin tests one week post-toxin. On day 14 we examined the behavioral response to injection of formalin into the hindpaw and found that responses during the second phase of pain behavior were significantly reduced. There was no change during the first phase. Formalin-evoked fos expression in the spinal cord was also reduced. We also evaluated morphine analgesia in the formalin test and found that toxin-treated animals exhibited enhanced morphine analgesia. These results establish the utility of using this immunotoxin to selectively destroy subpopulations of noradrenergic cell groups and provide evidence that acute and persistent nociception are differentially regulated by descending noradrenergic pathways.

Distinct biological subtypes and patterns of genome evolution in lymphoma revealed by circulating tumor DNA

Circulating tumor DNA reveals patterns of clonal evolution and allows classification of tumor subtypes in lymphoma. The telltale DNA in lymphoma Diffuse large B cell lymphoma is a relatively common type of tumor that can exhibit a wide range of behaviors, from indolent and curable cancers to ones that are very aggressive and difficult to treat. By analyzing DNA in tumor samples and blood of lymphoma patients, Scherer et al. have shown that specific genetic characteristics can determine each tumor’s cell of origin and identify tumors that are going to transform into more aggressive subtypes and may require more intensive treatment. The authors also demonstrated that circulating tumor DNA in the patients’ blood is suitable for this analysis, allowing for periodic monitoring of each patient without repeated invasive biopsies. Patients with diffuse large B cell lymphoma (DLBCL) exhibit marked diversity in tumor behavior and outcomes, yet the identification of poor-risk groups remains challenging. In addition, the biology underlying these differences is incompletely understood. We hypothesized that characterization of mutational heterogeneity and genomic evolution using circulating tumor DNA (ctDNA) profiling could reveal molecular determinants of adverse outcomes. To address this hypothesis, we applied cancer personalized profiling by deep sequencing (CAPP-Seq) analysis to tumor biopsies and cell-free DNA samples from 92 lymphoma patients and 24 healthy subjects. At diagnosis, the amount of ctDNA was found to strongly correlate with clinical indices and was independently predictive of patient outcomes. We demonstrate that ctDNA genotyping can classify transcriptionally defined tumor subtypes, including DLBCL cell of origin, directly from plasma. By simultaneously tracking multiple somatic mutations in ctDNA, our approach outperformed immunoglobulin sequencing and radiographic imaging for the detection of minimal residual disease and facilitated noninvasive identification of emergent resistance mutations to targeted therapies. In addition, we identified distinct patterns of clonal evolution distinguishing indolent follicular lymphomas from those that transformed into DLBCL, allowing for potential noninvasive prediction of histological transformation. Collectively, our results demonstrate that ctDNA analysis reveals biological factors that underlie lymphoma clinical outcomes and could facilitate individualized therapy.

How I treat CNS lymphomas

The pathogenesis of primary and secondary central nervous system (CNS) lymphoma poses a unique set of diagnostic, prognostic, and therapeutic challenges. During the past 10 years, there has been significant progress in the elucidation of the molecular properties of CNS lymphomas and their microenvironment, as well as evolution in the development of novel treatment strategies. Although a CNS lymphoma diagnosis was once assumed to be uniformly associated with a dismal prognosis, it is now reasonable to anticipate long-term survival, and possibly a cure, for a significant fraction of CNS lymphoma patients. The pathogenesis of CNS lymphomas affects multiple compartments within the neuroaxis, and proper treatment of the CNS lymphoma patient requires a multidisciplinary team with expertise not only in hematology/oncology but also in neurology, neuroradiology, neurosurgery, clinical neuropsychology, ophthalmology, pathology, and radiation oncology. Given the evolving principles of management and the evidence for improvements in survival, our goal is to provide an overview of current knowledge regarding the pathogenesis of CNS lymphomas and to highlight promising strategies that we believe to be most effective in establishing diagnosis, staging, and therapeutic management.

Outcome of patients with relapsed/refractory acquired immune deficiency syndrome-related lymphoma diagnosed 1999–2008 and treated with curative intent in the AIDS Malignancy Consortium

Abstract No comparative studies exist for relapsed/refractory (rel/rfr) acquired immune deficiency syndrome (AIDS)-related lymphoma (ARL). To determine practices over the last decade and to assess the outcomes of salvage chemotherapy with curative intent and autologous stem cell transplant (ASCT), we retrospectively evaluated treatment outcomes in patients with rel/rfr ARL who were treated in 13 national AIDS Malignancy Consortium (AMC) sites between 1999 and 2008 (n = 88). The most commonly used second-line therapies were ICE (ifosfamide/carboplatin/etoposide, n = 34), dose adjusted EPOCH (etoposide/prednisone/vincristine/cyclophosphamide/doxorubicin, n = 17) and ESHAP (etoposide/methylprednisolone/cytarabine/cisplatin, n = 11). The odds of achieving a response were lower for those with non-Hodgkin lymphoma (NHL) than for those with HL and for those with primary refractory disease than for those with relapse. Overall survival (OS) was significantly longer for those with relapsed disease compared to those with refractory disease and for those with non-Burkitt NHL compared to those with Burkitt. OS was longer in patients who underwent ASCT compared to those who did not (1-year OS: 63.2% vs. 37.2%). However, among 32 patients (36%) who achieved a complete or partial reponse (CR/PR) after second-line therapy, 1-year OS was not different between the two groups (87.5% for ASCT vs. 81.8% for non-ASCT). Long-term survival in some patients with rel/rfr ARL may be possible without transplant, although transplant remains the standard of care for chemotherapy sensitive disease.

Delayed hematopoietic recovery after auto-SCT in patients receiving arsenic trioxide-based therapy for acute promyelocytic leukemia: a multi-center analysis.

A potential link between arsenic (ATO)-based therapy and delayed hematopoietic recovery after autologous hematopoietic SCT (HSCT) for acute promyelocytic leukemia (APL) has previously been reported. We retrospectively reviewed the clinical histories of 58 patients undergoing autologous HSCT for APL at 21 institutions in the United States and Japan. Thirty-three (56%) of the patients received ATO-based therapy prior to stem cell collection. Delayed neutrophil engraftment occurred in 10 patients (17%): 9 of the 10 patients (90%) received prior ATO (representing 27% of all ATO-treated patients), compared with 1 of the 10 patients (10%) not previously treated with ATO (representing 4% of all ATO-naïve patients; P<0.001). Compared with ATO-naïve patients, ATO-treated patients experienced significantly longer times to ANC recovery (median 12 days vs 9 days, P<0.001). In multivariate analysis, the only significant independent predictor of delayed neutrophil engraftment was prior treatment with ATO (hazard ratio 4.87; P<0.001). Of the available stem cell aliquots from APL patients, the median viable post-thaw CD34+ cell recovery was significantly lower than that of cryopreserved autologous stem cell products from patients with non-APL AML. Our findings suggest that ATO exposure prior to CD34+ cell harvest has deleterious effects on hematopoietic recovery after autologous HSCT.

Long-term survival in AIDS-related primary central nervous system lymphoma

Background. The optimal therapeutic approach for patients with AIDS-related primary central nervous system lymphoma (AR-PCNSL) remains undefined. While its incidence declined substantially with combination antiretroviral therapy (cART), AR-PCNSL remains a highly aggressive neoplasm for which whole brain radiotherapy (WBRT) is considered a standard first-line intervention. Methods. To identify therapy-related factors associated with favorable survival, we first retrospectively analyzed outcomes of AR-PCNSL patients treated at San Francisco General Hospital, a public hospital with a long history of dedicated care for patients with HIV and AIDS-related malignancies. Results were validated in a retrospective, multicenter analysis that evaluated all newly diagnosed patients with AR-PCNSL treated with cART plus high-dose methotrexate (HD-MTX). Results. We provide evidence that CD4+ reconstitution with cART administered during HD-MTX correlates with long-term survival among patients with CD4 <100. This was confirmed in a multicenter analysis which demonstrated that integration of cART regimens with HD-MTX was generally well tolerated and resulted in longer progression-free survival than other treatments. No profound differences in immunophenotype were identified in an analysis of AR-PCNSL tumors that arose in the pre- versus post-cART eras. However, we detected evidence for a demographic shift, as the proportion of minority patients with AR-PCNSL increased since advent of cART. Conclusion. Long-term disease-free survival can be achieved in AR-PCNSL, even among those with histories of opportunistic infections, limited access to health care, and medical non-adherence. Given this, as well as the long-term toxicities of WBRT, we recommend that integration of cART plus first-line HD-MTX be considered for all patients with AR-PCNSL.

The transfusion tether: Bridging the gap between end-stage hematologic malignancies and optimal end-of-life care

Two patients with refractory acute myeloid leukemia (AML) were recently treated at our institution. Patient 1 was a fit 72-year-old gentleman who failed to remit after high-intensity induction chemotherapy. Despite a second induction, his leukemia persisted. Although his functional status remained excellent, he eventually developed platelet-directed antibodies and required outpatient transfusion of HLA-matched platelets several times per week. When his leukemia eventually progressed, he suffered a stroke, leaving him unable to care for himself at home. Reluctant to forego lifeprolonging transfusion support, he refused to be discharged from the hospital unless he could continue to receive a similar level of transfusion support as he had been receiving. He remained hospitalized for 36 days while awaiting identification of a facility that was willing to continue transfusions. Patient 2 was a 79-year-old gentleman whose leukemia had been controlled for more than a year on various low-intensity, outpatient-based regimens. Like Patient 1, he too developed progressive transfusion dependence, requiring either blood and/or platelets multiple times per week. He had a long history of cardiovascular disease, and severe anemia typically precipitated dyspnea and angina, both of which resolved with transfusion. Unfortunately, his performance status steadily declined, and at his final clinic visit he felt so poorly that he requested to be “euthanized.” He was transferred from clinic directly to an inpatient hospice unit, where he died less than 24 hours later. Patients 1 and 2 illustrate one of the unique challenges faced by the practitioners who care for patients with advanced hematologic malignancies—the “transfusion tether.” Unlike many costly “breakthrough” treatments for solid tumors—typically lauded for providing modest gains in progression-free survival—many patients with terminal hematologic malignancies can live equally long and can feel better with intensive transfusion support alone. While recent studies suggest a benefit in terms of both quality of life and even overall survival in patients with solid tumors who receive early integration of palliative care, palliative care and hospice utilization for patients with hematologic malignancies remains low [1,2]. This is, in part, because no such data are yet available specific to these patients. Nonetheless, it seems reasonable to extrapolate that improvements in symptom management and end-of-life transitions might translate into similar gains for patients with liquid tumors. Interestingly, however, the very benefits afforded by transfusion support can paradoxically prevent patients with end-stage hematologic malignancies from receiving the more comprehensive palliative support provided by hospice care. A recent meta-analysis demonstrated that patients with hematologic malignancies were significantly less likely to receive specialist palliative care or hospice services compared to patients with solid tumors (RR 0.46, 95% CI 0.42–0.50) [3]. In another retrospective cohort of nearly 350 patients referred to the palliative care program at M.D. Anderson Cancer Center, the median time from palliative care referral to death was 0.6 months for patients with hematologic malignancies versus 2 months for patients with solid tumors (P< 0.001) [4]. At Massachusetts General Hospital and the Dana-Farber Cancer Institute, end-of-life health care utilization was specifically studied in older patients with AML, with only 47 of 330 patients (14.2%) ever receiving a palliative care consultation, and a median time from palliative care consultation to death of 7 days [5]. In our practice, reluctance to discontinue transfusion support at the end of life—whether patient-driven, physician-driven, or both—is one of the principal barriers to optimizing end-of-life care for our patients. While not explicitly contraindicated in the eligibility criteria for Medicare’s hospice benefit, hospice organizations frequently opt to exclude patients who wish to receive ongoing transfusions. A survey of 591 hospices in the United States found that 40% of hospices refused to allow transfusion support for their patients [6]. It is likely that a much higher percentage of hospices would have excluded patients with end-stage hematologic malignancies, whose transfusion needs are typically much higher than for the average terminally ill patient. The “palliative” benefit of transfusion support to reduce dyspnea, fatigue, bleeding, and weakness is relatively well-documented [7]. But in addition to receiving symptomatic relief, it is clear that many patients who continue transfusions also derive benefit from the regular interactions with the medical assistants, nurses, and physicians with whom they have formed strong personal connections over the course of their treatment. When patients cease to be able to receive transfusions, this network of support is severed. The obvious and not inconsequential impact of intensive transfusion support for patients with terminal disease is the financial burden placed on the healthcare system. Medicare hospice benefits pay a fixed, per patient, per day benefit to hospice organizations, regardless of the services

Plasmablastic lymphoma is treatable in the HAART era. A 10 year retrospective by the AIDS Malignancy Consortium

Plasmablastic lymphoma (PBL) is an aggressive CD20 negative diffuse large B cell lymphoma over-represented in patients with HIV infection. CD45 expression is weak, but immunoglobulin genes are rearranged and plasma cell markers are expressed.[1] In 1997, 95% of the cases were reported to be from HIV + patients, all with oral cavity involvement, 68% stage I. In this pre-HAART era report, 9//11 patients with follow-up died within a year of diagnosis. Most other case reports and case series have typically described poor survival for patients with PBL, particularly those infected with HIV,[2] while others were more optimistic.[3] We sought to characterize patients with PBL diagnosed and treated solely in the HAART era. We identified 12 patients with newly diagnosed PBL treated at the AIDS Malignancy Consortium (AMC) sites from 1999 to 2008. This retrospective analysis suggests these patients had better outcomes than those identified pre-HAART, perhaps due to use of aggressive chemotherapy made possible because of better supportive care and antiretroviral therapy. All AMC sites, which participated in this retrospective review were queried for cases of PBL diagnosed from 1998–2008. Two of the authors (AC and AN) reviewed the pathology reports for the criteria for plasmablastic lymphoma described in the 2008 WHO Classification.[4] Twelve cases from 9 AMC sites were included in this study. Descriptive statistics were computed for demographic and clinical characteristics. Overall survival (OS) was calculated from date of initial diagnosis to death or last follow-up. Kaplan-Meier estimates of 1-year survival were computed. All AMC sites had an Institutional Review Board waiver of authorization. Baseline clinical characteristics at study entry are presented in Table 1. The median CD4 + count at HIV diagnosis was 256 cells/uL (range 45–750) and was lower at initial PBL diagnosis with a median of 136 cells/uL (range of 2–514). Sixty-seven percent of the patients had had a prior opportunistic infection. Most (58%) of patients were not on HAART at lymphoma diagnosis, however, they had all previously taken HAART at some point. Of 7 patients not on HAART, 6 started HAART, typically at diagnosis or chemoimmunotherapy initiation. Stage at initial diagnosis was I (25%), II (25%), III (0%) and IV (50%). Four of 7 patients with extranodal disease had more than one site of involvement. Extranodal sites of disease at initial diagnosis included bone without bone marrow (4), bone marrow (1), liver (2), kidney (2), sinus (1), cerebrospinal fluid (1), colon (1), skin (1), adrenal (1), nasopharynx (1) and stomach (1). Table 1 Clinical characteristics at study entry of 12 HIV-positive patients with initial diagnosis of plasmablastic lymphoma. Surprisingly, no patients had oral involvement. LDH was elevated in 5/8 where the value was known. The International Prognostic Index could not be calculated for the group as a whole as performance status assessment data was not available in one third of the patients. Not all cases had uniform immunophenotypic data available [Table 1]. As per the definition of plasmablastic lymphoma, all 12 cases tested were negative for the B cell marker CD20. Similarly, markers of terminal B cell differentiation, CD138 and MUM-1/IFR4, were positive in 6/6 cases and in 4/4 cases tested, respectively Epstein-Barr virus (EBV) was present in 8/8 cases based on in situ hybridization (EBER). At initial diagnosis, 10/12 patients received chemotherapy, although HAART alone was attempted without success in one patient. Treatment was CHOP on a 14 day cycle (n=1) [5] or 21 day cycle (n=3), [6] (cyclophosphamide, doxorubicin, vincristine, prednisone), infusional CDE (n=1), (cyclophosphamide, doxorubicin, etoposide); [7] infusional EPOCH (n=2) (cyclophosphamide, doxorubicin, vincristine, etoposide and prednisone),[8,9] or other (n=5). The other therapies included EPOCH with high dose methotrexate and zidovudine, either alternating (n=2), or sequential (n=2). Three patients with stage I/II disease received radiation in combination with chemotherapy. Two of the ten treated patients experienced grade 3/4 toxicity. No patient died of treatment. One patient experienced grade 3/4 fatigue, anemia, thrombocytopenia, febrile neutropenia, nausea, vomiting, diarrhea, and weight loss, and the other patient experienced renal insufficiency. Responses were complete (CR) in 7, partial (PR) in 2 and refractory in 1. CRs were seen with CHOP (n=4), EPOCH (n=2), and EPOCH alternating with high dose methotrexate and zidovudine (n=1). PRs were seen after EPOCH alternating with high dose methotrexate and zidovudine (n=2). The one patient treated with CDE had refractory disease. Overall survival is shown in Fig. 1. At a median follow up of 73 weeks (range, 40–165), the median survival was not reached. The one-year survival was 66.7% (SE, 13.6). No patients died in the follow up period after year one. Figure 1 Survival of plasmablastic lymphoma patients. We report the first case series of plasmablastic lymphoma patients, under the care of dedicated HIV malignancy oncologists in a consortium setting, diagnosed and treated exclusively in the HAART era. In this study we demonstrate relatively long disease free survival despite earlier reports showing nearly all HIV + patients with PBL were destined to die from progressive lymphoma. In the initial, pre-HAART study of PBL by Delecluse, et al.[1] 9 of the 11 patients died within 16 months of diagnosis including 3 patients with stage I disease treated with external beam radiation, poly-chemotherapy or a combination of both. In the HAART era, intensive treatment of aggressive lymphomas, such as Burkitt and diffuse large B cell lymphoma, can result in CR and long-term survival for HIV infected patients.[9–11] Cohort studies suggest that plasmablastic lymphoma has an inferior survival to Burkitt and DLBCL.[12] We demonstrate that PBL may be curable, even in patients with higher stage disease and more extranodal involvement compared to the cases described by Delecluse, et al. where the disease was primarily localized to the jaw in the majority of patients. Recently, Ibrahim et al. reported in abstract form, their single institution experience with 25 patients during the HAART era.[13] Similar to our study, only 32% were on HAART and the median CD4 + cell count was 87cells/uL. Only 76% received chemotherapy, with EPOCH being the most common treatment choice (56%), and 28% received radiotherapy. The response rate was 84% and the median OS observed in all patients who received chemotherapy was 11.6 months (range, 2–63 months). Those who were treated with EPOCH demonstrated a better median overall survival (17 months) compared to those treated with CHOP and CHOP-like regimens (7 months, p=0.04). Castillo et al. reviewed 112 cases in the literature and noted a median overall survival of 15 months with a 3 year OS of only 25%. In contrast, Cattaneo et al. reported their single institution experience of 13 patients treated during the HAART era with a 67% 3 year OS.[14] 15 patients were treated, all with CHOP or CHOP-like regiments, with involved field consolidation in seven. Five received high dose therapy with autologus stem cell transplant as first CR consolidation. The initial goal of this AMC retrospective study was to identify which first line therapy might be the most promising for HIV positive patients with plasmablastic lymphoma. The rarity of this lymphoma resulted in a very small cohort, possibly as a result of incomplete retrieval from institutional databases. In this regard, we were unable to draw specific conclusions due to the sample size, the disparate number of lymphoma regimens, and the retrospective nature of the study. However, it is notable that CR was achieved with CHOP or EPOCH based regimens. The literature on regimens more intensive than CHOP is inconsistent. Two earlier retrospective studies did not demonstrate an advantage of therapies more intensive than CHOP for PBL.[15,16] Loghavi et al. noted a series of 50 patients (20 HIV+) tended to have a better OS with CHOP than hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone (hyper-CVAD) (p=0.078).[17] Nonetheless, many experts do consider CHOP inadequate and favor intensive therapies including EPOCH.[18] As an aside, we cannot comment on the efficacy of HAART therapy alone, as only one patient was treated with this strategy with disease progression. In addition to intensive chemotherapy, it is probable that our patients fared better than those in the pre- HAART era due in part to better supportive care and more robust immune status. However, we cannot be certain, as the median CD4 + cell count of 256 cells/uL at initial PBL diagnosis was relatively low. Moreover, 59% had had a prior opportunistic infection and only 29% were on HAART at PBL diagnosis. Currently Clinical Trials Support Unit (CTSU) 9177 is prospectively studying EPOCH therapy for patients diagnosed with PBL irrespective of HIV status. Given the biologic similarity to multiple myeloma, drug classes like immune modulatory agents (IMiDs) and protease inhibitors [19] might also be promising agents for future study. The nearly ubiquitous finding of Epstein Barr in tumor tissue (11of 12 evaluable cases in our study) may provide a rationale for viral-based strategies. Finally, the role for autologous stem cell transplant in the setting of relapsed/refractory disease is also worthy of exploration.

Fourth complete remission with immunosuppression withdrawal and irinotecan after both autologous and allogeneic transplants for diffuse large B cell lymphoma.

Diffuse large B cell lymphoma (DLBCL) relapsing a second time after high-dose therapy and autologous transplantation is rarely cured. Allogeneic transplant has been curative in lymphoma [1], but the true applicability has not been fully determined. Treatment of a third relapse after sequential autologous and allogeneic transplants would not be expected to result in a durable remission. We present here a patient with multiply-relapsed, large B cell nonHodgkin lymphoma (NHL) after both autologous transplant in second remission and allogeneic transplant in third remission. She is now in an exceptional sustained fourth complete remission after withdrawal of immunosuppression and treatment with irinotecan. A 29-year-old Taiwanese woman presented in 1999 with several weeks of right sided facial swelling and ipsilateral supraclavicular adenopathy unresponsive to antibiotics. A submandibular lymph node biopsy revealed follicular and diffuse large cleaved cell lymphoma. Computed tomography (CT), fluorine-18, 2-fluoro-2-deoxy-D-glucose positron emission tomography (FDG PET), and bone marrow biopsy demonstrated stage IIA disease. She had no International Prognostic Index (IPI) risk factors. She received an abbreviated, three-cycle regimen of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) followed by radiation localized to the right neck and supraclavicular region (180 cGy617 fractions, for a total of 3060 Gy) [2]. She tolerated the treatments well and follow-up imaging revealed no evidence of disease. The patient remained disease-free for 7 years, including 5 years of radiographic surveillance. Seven years after combined modality therapy, she presented with a new, left-sided neck mass. Excisional biopsy demonstrated DLBCL with plasmacytoid differentiation. Clinical stage IIIS disease was established by CT, FDG PET, and bone marrow biopsy with two adverse IPI risk factors. She received an institutional review board approved protocol for cytoreduction with two cycles of rituximab combined with augmented ifosfamide, carboplatin, and etoposide (R-ICE) [ifosfamide 10 000 mg/m over 2 days and etoposide 200 mg/m Q12 h63 doses with carboplatinum at an area under the curve of 5]. Doses were augmented compared with our original report of ICE as second-line therapy before autologous stem cell transplantation (ASCT) for primary refractory or relapsed DLBCL [3]. Follow-up CT and PET-imaging did not show any residual disease. The stem cell harvest yielded 17.6610 CD34þ cells. Autologous transplant with 7.0610 CD34þ cells/kg followed conditioning with carmustine, etoposide, cytarabine, and melphalan (BEAM). She also received a single dose of rituximab 1 month after autologous transplantation. Approximately 7 months after ASCT, the patient noticed new left cervical adenopathy coupled with fevers and progressive jaundice in the preceding

Circulating Tumor DNA Measurements As Early Outcome Predictors in Diffuse Large B-Cell Lymphoma

PURPOSE Outcomes for patients with diffuse large B-cell lymphoma remain heterogeneous, with existing methods failing to consistently predict treatment failure. We examined the additional prognostic value of circulating tumor DNA (ctDNA) before and during therapy for predicting patient outcomes. PATIENTS AND METHODS We studied the dynamics of ctDNA from 217 patients treated at six centers, using a training and validation framework. We densely characterized early ctDNA dynamics during therapy using cancer personalized profiling by deep sequencing to define response-associated thresholds within a discovery set. These thresholds were assessed in two independent validation sets. Finally, we assessed the prognostic value of ctDNA in the context of established risk factors, including the International Prognostic Index and interim positron emission tomography/computed tomography scans. RESULTS Before therapy, ctDNA was detectable in 98% of patients; pretreatment levels were prognostic in both front-line and salvage settings. In the discovery set, ctDNA levels changed rapidly, with a 2-log decrease after one cycle (early molecular response [EMR]) and a 2.5-log decrease after two cycles (major molecular response [MMR]) stratifying outcomes. In the first validation set, patients receiving front-line therapy achieving EMR or MMR had superior outcomes at 24 months (EMR: EFS, 83% v 50%; P = .0015; MMR: EFS, 82% v 46%; P < .001). EMR also predicted superior 24-month outcomes in patients receiving salvage therapy in the first validation set (EFS, 100% v 13%; P = .011). The prognostic value of EMR and MMR was further confirmed in the second validation set. In multivariable analyses including International Prognostic Index and interim positron emission tomography/computed tomography scans across both cohorts, molecular response was independently prognostic of outcomes, including event-free and overall survival. CONCLUSION Pretreatment ctDNA levels and molecular responses are independently prognostic of outcomes in aggressive lymphomas. These risk factors could potentially guide future personalized risk-directed approaches.