scholarly journals A Clinical Prognostic Model Based on Machine Learning from the Fondazione Italiana Linfomi (FIL) MCL0208 Phase III Trial

Cancers ◽  
2021 ◽  
Vol 14 (1) ◽  
pp. 188
Author(s):  
Gian Maria Zaccaria ◽  
Simone Ferrero ◽  
Eva Hoster ◽  
Roberto Passera ◽  
Andrea Evangelista ◽  
...  

Background: Multicenter clinical trials are producing growing amounts of clinical data. Machine Learning (ML) might facilitate the discovery of novel tools for prognostication and disease-stratification. Taking advantage of a systematic collection of multiple variables, we developed a model derived from data collected on 300 patients with mantle cell lymphoma (MCL) from the Fondazione Italiana Linfomi-MCL0208 phase III trial (NCT02354313). Methods: We developed a score with a clustering algorithm applied to clinical variables. The candidate score was correlated to overall survival (OS) and validated in two independent data series from the European MCL Network (NCT00209222, NCT00209209); Results: Three groups of patients were significantly discriminated: Low, Intermediate (Int), and High risk (High). Seven discriminants were identified by a feature reduction approach: albumin, Ki-67, lactate dehydrogenase, lymphocytes, platelets, bone marrow infiltration, and B-symptoms. Accordingly, patients in the Int and High groups had shorter OS rates than those in the Low and Int groups, respectively (Int→Low, HR: 3.1, 95% CI: 1.0–9.6; High→Int, HR: 2.3, 95% CI: 1.5–4.7). Based on the 7 markers, we defined the engineered MCL international prognostic index (eMIPI), which was validated and confirmed in two independent cohorts; Conclusions: We developed and validated a ML-based prognostic model for MCL. Even when currently limited to baseline predictors, our approach has high scalability potential.

Haematologica ◽  
2019 ◽  
Vol 105 (11) ◽  
pp. 2598-2607 ◽  
Author(s):  
Sonia Jaramillo ◽  
Andreas Agathangelidis ◽  
Christof Schneider ◽  
Jasmin Bahlo ◽  
Sandra Robrecht ◽  
...  

Almost one-third of all patients with chronic lymphocytic leukemia (CLL) express stereotyped B cell receptor immunoglobulins (BcR IG) and can be assigned to distinct subsets, each with a particular BcR IG. The largest stereotyped subsets are #1, #2, #4 and #8, associated with specific clinicobiological characteristics and outcomes in retrospective studies. We assessed the associations and prognostic value of these BcR IG in prospective multicenter clinical trials reflective of two different clinical situations: i) early-stage patients (watch-and-wait arm of the CLL1 trial) (n=592); ii) patients in need of treatment, enrolled in 3 phase III trials (CLL8, CLL10, CLL11), treated with different chemo-immunotherapies (n=1861). Subset #1 was associated with del(11q), higher CLL international prognostic index (CLL-IPI) scores and similar clinical course to CLL with unmutated immunoglobulin heavy variable (IGHV) genes (U-CLL) in both early and advanced stage groups. IGHV-mutated (M-CLL) subset #2 cases had shorter time-to-first-treatment (TTFT) versus other M-CLL cases in the early-stage cohort (HR: 4.2, CI: 2-8.6, p<0.001), and shorter time-to-next-treatment (TTNT) in the advanced-stage cohort (HR: 2, CI: 1.2-3.3, p=0.005). M-CLL subset #4 was associated with lower CLL-IPI scores and younger age at diagnosis; in both cohorts, these patients showed a trend towards better outcomes versus other M-CLL. U-CLL subset #8 was associated with trisomy 12. Overall, this study shows that major stereotyped subsets have distinctive characteristics. For the first time in prospective multicenter clinical trials, subset # 2 appeared as an independent prognostic factor for earlier TTFT and TTNT and should be proposed for risk stratification of patients.


2008 ◽  
Vol 26 (28) ◽  
pp. 4579-4586 ◽  
Author(s):  
Robert Marcus ◽  
Kevin Imrie ◽  
Philippe Solal-Celigny ◽  
John V. Catalano ◽  
Anna Dmoszynska ◽  
...  

PurposeTo compare the long-term outcome of patients with previously untreated follicular lymphoma (FL) needing therapy, after treatment with cyclophosphamide, vincristine and prednisone (CVP) versus CVP plus rituximab (R-CVP) and to evaluate the predictive value of known prognostic factors after treatment with R-CVP.Patients and MethodsPatients with previously untreated CD20-positive stage III/IV FL were randomly assigned to eight cycles of R-CVP (n = 159) or CVP alone (n = 162). The median follow-up period was 53 months.ResultsThe primary end point—time to treatment failure (TTF), which included patients without a response after four cycles as an event—was significantly prolonged in patients receiving R-CVP versus CVP (P < .0001). Improvements in all other end points, including overall and complete response rates (P < .0001), time to progression (TTP; P < .0001), response duration (P < .0001), time to next antilymphoma treatment (P < .0001), and overall survival (OS; P = .029; 4-year OS: 83% v 77%;) were achieved with R-CVP versus CVP alone. Univariate analyses demonstrated an improvement in TTP with R-CVP versus CVP irrespective of the Follicular Lymphoma International Prognostic Index (FLIPI) subgroup, the International Prognostic Index (IPI) subgroup, baseline histology, and the presence or absence of B symptoms or bulky disease. By multivariate analysis, FLIPI retains a strong predictive power for TTP in the presence of the trial treatment effect.ConclusionAnalysis of all outcome measures, including OS, confirm the benefit of adding R to CVP in the front-line treatment of FL.


Blood ◽  
2005 ◽  
Vol 106 (11) ◽  
pp. 4740-4740
Author(s):  
Athanasios G. Galanopoulos ◽  
Anastasia Tsakiridou ◽  
Eurydiki Michalis ◽  
Theodoros Marinakis ◽  
George Gortzolidis ◽  
...  

Abstract Background: The treatment of patients with chronic lymphocytic leukemia (CLL) with Rituximab in combination with fludarabine and cyclophosphamide was reported to be more efficacious, in terms of complete and molecular remission compared with historical data for fludarabine plus cyclophosphamide (S.O’Brien, Haematologica2002; 87:50–53). Aims: Evaluation of the clinical efficacy and toxicity of the FCR combination in patients of our Haematologic Centre. Methods: Seventeen patients, 8 males and 9 females with a median age of 69,5 years, with relapsed/refractory or de novo CLPD (9 CLL and 8 NHL patients) were enrolled in this study between February 2002 and August 2004. Fifty percent of CLL patients had Rai stage I/II and the rest 50% had Rai stage III/IV disease. Four NHL patients had an International Prognostic Index (IPI) 2, one patient IPI 3 and three patients IPI 4. All patients were treated with Rituximab 375 mg/m2 on day1 in combination with Fludarabine and Cyclophosphamide (25 mg/m2 and 250 mg/m2 respectively) for days 2 to 4, every 4 weeks, for 6 consecutive cycles. Nine patients had a history of a prior unsuccessful treatment. Results: Overall, 14 out of 17 evaluable patients (82%) were responsive to the treatment [12 patients (70%) complete response (CR) and 2 patients (12%) partial response (PR)]. The remaining 3 patients had progressive disease (NR). Hematological toxicity was acceptable (grade 2–3 neutropenia in 6/17 patients, grade 2–3 anemia and thrombocytopenia in 2/17 patients). There were no septic episodes except one case with neutropenic fever. There were no adverse events like nausea or vomiting except one patient with a serious anaphylactic reaction due to Rituximab administration. Three CLL patients died because of progressive disease. Summary/conclusions: this preliminary report suggests that the FCR regimen is an effective and safe treatment for CLPD patients, achieving higher CR rates than previous treatments. A longer follow up of a larger number of patients is required to confirm an improved survival in these patients.


2012 ◽  
Vol 30 (15_suppl) ◽  
pp. 4644-4644 ◽  
Author(s):  
Guru Sonpavde ◽  
Gregory Russell Pond ◽  
Stephen John Clarke ◽  
Janette L. Vardy ◽  
S. L. Wang ◽  
...  

4644 Background: A prognostic model for mCRPC post docetaxel is necessary to guide therapy. We retrospectively analyzed a phase III trial enrolling progressive mCRPC following docetaxel to construct a prognostic model. Additionally, we studied the impact of neutrophil-lymphocyte ratio (NLR), a potential marker for inflammatory and immune state. Methods: A phase III trial (SUN-1120) comparing prednisone combined with sunitinib (N=584) or placebo (N=289) for mCRPC following docetaxel-based chemotherapy was evaluated. The treatment arms were combined for analysis, since no statistical difference was observed in the primary endpoint of overall survival (OS). A logarithmic transformation was applied to non-normal factors. The Kaplan-Meier method was used for OS estimation. To identify an optimal prognostic model for survival, we used a Cox proportional hazards regression methods with forward stepwise selection, stratifying for ECOG PS, progression type (PSA or radiographic) and treatment group. A risk score was calculated and patients were categorized into risk groups to assess model performance. Results: Data from patients without missing data (n=806) were used to construct an optimal model. The factors used in the model that remained individually significant in multivariate analysis were: log-LDH (HR 2.77 [95% CI=2.23, 3.44], p<0.001), hemoglobin (0.81 [0.76, 0.87], p<0.001), log-NLR (1.63 [1.38, 1.92], p<0.001), >1 organ involved (1.53 [1.24, 1.88], p<0.001), log-alkaline phosphatase (1.14 [1.01, 1.30], p=0.041) and log-PSA (1.07 [1.00, 1.13], p=0.036). No clear cutpoints were identified; thus, these prognostic factors were used to group patients into 3 equally sized risk categories. Low, medium and high risk patients (n=268-270 per group) had median (95% CI) OS estimates of 23.7 (21.4-not reached), 13.5 (11.6-15.8) and 7.3 (6.3-8.4) months, respectively. Conclusions: A prognostic risk model with readily available variables significantly discriminated between outcomes in post-docetaxel mCRPC and may provide valuable information in future studies. High NLR was associated with an independent poor prognostic impact, and warrants prospective validation.


2009 ◽  
Vol 27 (10) ◽  
pp. 1607-1614 ◽  
Author(s):  
Howard Hochster ◽  
Edie Weller ◽  
Randy D. Gascoyne ◽  
Thomas M. Habermann ◽  
Leo I. Gordon ◽  
...  

Purpose To determine if maintenance rituximab (MR) after standard chemotherapy improves progression-free survival (PFS) in advanced-stage indolent lymphoma. Patients and Methods Patients with stage III-IV indolent lymphoma with responding or stable disease after cyclophosphamide, vincristine, and prednisone (CVP) chemotherapy were stratified by initial tumor burden, residual disease after CVP (minimal or gross), and histology, and randomly assigned to observation (OBS) or MR 375 mg/m2 once per week for 4 weeks every 6 months for 2 years. PFS was the primary end point. Results Three hundred eleven (282 with follicular lymphoma) evaluable patients who received CVP were randomly assigned to OBS (n = 158) or MR (n = 153). Best response improved in 22% MR versus 7% OBS patients (P = .00006). Toxicity was minimal in both study arms. Three-year PFS after random assignment was 68% MR versus 33% OBS (hazard ratio [HR] = 0.4; P = 4.4 × 10−10 [all patients]) and 64% MR v 33% OBS (HR = 0.4; P = 9.2 × 10−8 [patients with follicular lymphoma]). There was an advantage for MR regardless of Follicular Lymphoma International Prognostic Index score, tumor burden, residual disease, or histology. In multivariate analysis of MR patients, minimal disease after CVP was a favorable prognostic factor. OS at 3 years was 92% MR versus 86% OBS (HR = 0.6; log-rank one-sided P = .05) and, among patients with follicular lymphoma, OS was 91% MR versus 86% (HR = 0.6; log-rank one-sided P = .08). A trend favoring MR was observed among patients with high tumor burden (log-rank one-sided P = .03). Conclusion The E1496 study provides the first phase III data in untreated indolent lymphoma that MR after chemotherapy significantly prolongs PFS.


2014 ◽  
Vol 32 (7) ◽  
pp. 671-677 ◽  
Author(s):  
Susan Halabi ◽  
Chen-Yen Lin ◽  
W. Kevin Kelly ◽  
Karim S. Fizazi ◽  
Judd W. Moul ◽  
...  

Purpose Prognostic models for overall survival (OS) for patients with metastatic castration-resistant prostate cancer (mCRPC) are dated and do not reflect significant advances in treatment options available for these patients. This work developed and validated an updated prognostic model to predict OS in patients receiving first-line chemotherapy. Methods Data from a phase III trial of 1,050 patients with mCRPC were used (Cancer and Leukemia Group B CALGB-90401 [Alliance]). The data were randomly split into training and testing sets. A separate phase III trial served as an independent validation set. Adaptive least absolute shrinkage and selection operator selected eight factors prognostic for OS. A predictive score was computed from the regression coefficients and used to classify patients into low- and high-risk groups. The model was assessed for its predictive accuracy using the time-dependent area under the curve (tAUC). Results The model included Eastern Cooperative Oncology Group performance status, disease site, lactate dehydrogenase, opioid analgesic use, albumin, hemoglobin, prostate-specific antigen, and alkaline phosphatase. Median OS values in the high- and low-risk groups, respectively, in the testing set were 17 and 30 months (hazard ratio [HR], 2.2; P < .001); in the validation set they were 14 and 26 months (HR, 2.9; P < .001). The tAUCs were 0.73 (95% CI, 0.70 to 0.73) and 0.76 (95% CI, 0.72 to 0.76) in the testing and validation sets, respectively. Conclusion An updated prognostic model for OS in patients with mCRPC receiving first-line chemotherapy was developed and validated on an external set. This model can be used to predict OS, as well as to better select patients to participate in trials on the basis of their prognosis.


2013 ◽  
Vol 31 (23) ◽  
pp. 2903-2911 ◽  
Author(s):  
Rashmi S. Goswami ◽  
Eshetu G. Atenafu ◽  
Yali Xuan ◽  
Levi Waldron ◽  
Patricia P. Reis ◽  
...  

Purpose Mantle-cell lymphoma (MCL) has a variable natural history but is incurable with current therapies. MicroRNAs (miRs) are useful in prognostic assessment of cancer. We determined an miR signature defining aggressiveness in B-cell non-Hodgkin lymphomas (NHL) and assessed whether this signature aids in MCL prognosis. Methods We assessed miR expression in a training set of 43 NHL cases. The miR signature was validated in 44 additional cases and examined on a training set of 119 MCL cases from four institutions in Canada. miRs significantly associated with overall survival were examined in an independent cohort of 114 MCL cases to determine association with patient outcome. miR expression was combined with current clinical prognostic factors to develop an enhanced prognostic model in patients with MCL. Results Fourteen miRs were differentially expressed between aggressive and indolent NHL; 11 of 14 were validated in an independent set of NHL (excluding MCL). miR-127-3p and miR-615-3p were significantly associated with overall survival in the MCL training set. Their expression was validated in an independent MCL patient set. In comparison with Ki-67, expression of these miRs was more significantly associated with overall survival among patients with MCL. miR-127-3p was combined with Ki-67 to create a new prognostic model for MCL. A similar model was created with miR-615-3p and Mantle Cell Lymphoma International Prognostic Index scores. Conclusion Eleven miRs are differentially expressed between aggressive and indolent NHL. Two novel miRs were associated with overall survival in MCL and were combined with clinical prognostic models to generate novel prognostic data for patients with MCL.


2009 ◽  
Vol 27 (18_suppl) ◽  
pp. 2-2 ◽  
Author(s):  
S. J. Schuster ◽  
S. S. Neelapu ◽  
B. L. Gause ◽  
F. M. Muggia ◽  
J. P. Gockerman ◽  
...  

2 Background: In previous trials, tumor-specific purified idiotype (Id) protein conjugated to keyhole limpet hemocyanin (KLH) administered with granulocyte-monocyte colony-stimulating factor (GM-CSF) induced follicular lymphoma (FL)-specific immune responses and molecular remissions (Nat Med. 1999;5:1171–7). Methods: We conducted a prospective randomized double-blind placebo-controlled multicenter phase III study of patient-specific autologous tumor-derived Id vaccine in advanced stage previously untreated FL patients (pts) with a lymph node adequate for vaccine production (≥ 2cm). Pts achieving complete response (CR) or complete response unconfirmed (CRu) after chemotherapy (PACE: prednisone, doxorubicin, cyclophosphamide, etoposide) were stratified by International Prognostic Index risk group and randomized 2:1 to receive either vaccination with Id-KLH/GM-CSF or control (KLH/GM-CSF). The primary endpoint was disease free survival. Results: 234 pts were enrolled; 177 (76%) achieved CR/CRu and were randomized. Of 177 randomized pts, 117 maintained CR/CRu ≥ 6 mo per protocol requirement and then received at least one dose of vaccine, 55 relapsed before vaccination, 4 were vaccine manufacturing failures, and 1 violated protocol. Pts who received ≥ one vaccine dose constituted the modified intent-to-treat population for determination of efficacy. 76 pts received Id-KLH/GM-CSF and 41 pts received the control (KHL/GM-CSF). No serious adverse events were attributed to Id vaccination. At a median follow-up of 56.6 mo (range 12.6 –89.3 mo), median time to relapse after randomization for the Id-KLH/GM-CSF arm was 44.2 mo, versus 30.6 mo for the control arm (p = 0.045; HR = 1.6). Conclusions: Id vaccination after a chemotherapy-induced remission of ≥ 6 mo prolongs remission duration in pts with FL. Compared to other phase III Id vaccine trials, the positive outcome of this study may reflect application of Id vaccine in pts in CR/CRu or use of hybridomas to produce Id. Genomic and immune response analyses are planned on residual autologous tumor and blood samples. Additional studies of this patient-specific vaccine in FL pts pretreated with anti-CD20 antibody-containing chemotherapy are indicated. [Table: see text]


2021 ◽  
Vol 39 (15_suppl) ◽  
pp. TPS7567-TPS7567
Author(s):  
Deborah Marie Stephens ◽  
Anna Moseley ◽  
Brian T. Hill ◽  
John M. Pagel ◽  
Mazyar Shadman ◽  
...  

TPS7567 Background: Currently, asymptomatic patients with CLL/SLL are observed without treatment until development of symptoms or cytopenias. Historically, early intervention studies with chemoimmunotherapy have not resulted in an overall survival (OS) benefit and have resulted in toxicity. The introduction of targeted therapies, such as venetoclax and obinutuzumab (VO), have provided tolerable/efficacious options for CLL patients. In the CLL14 study, symptomatic CLL patients receiving frontline therapy with VO had longer progression-free survival (PFS) and deeper remissions [more minimal residual disease-undetectable (MRDu)] compared with those receiving chlorambucil and obinutuzumab (Fischer 2019). The CLL-International Prognostic Index (CLL-IPI; Table) is a validated prognostic model to predict which patients are highest risk for a shorter time to first therapy and shorter OS. We aim to use VO as early intervention in asymptomatic, high-risk patients with CLL to potentially lengthen OS and thus alter the natural history of the disease. Methods: On 12/14/20, we activated the S1925 study for adult patients with CLL or SLL, who were diagnosed within 12 months of enrollment. Eligible patients have a CLL-IPI score ≥ 4 (Table) or complex cytogenetics (≥3 cytogenetic abnormalities) and do not meet any criteria for initiation of treatment by the International Working Group for CLL (IWCLL; Hallek 2018) guidelines. Enrolled patients are randomized in a 2:1 manner to early versus delayed (at the time IWCLL indication for treatment is met) therapy with VO. VO is administered for a fixed duration of 12 months as previously described (Fischer 2019). The primary endpoint is OS. We hypothesize that early intervention with VO will improve the rate of 6-year OS from 60% to 80%. This design requires 222 eligible patients for 88% power (2-sided α=0.05) for the primary comparison. To allow for 10% ineligibility, we will enroll 247 patients. Estimated accrual time is 4 years. Secondary endpoints include: rates of response, PFS, and relapse-free survival; safety; time to 2nd CLL-directed therapy; and quality of life (FACT-Leukemia total score). The primary translational objective is to evaluate the prognostic association between OS and peripheral blood MRD status at 15 months after treatment initiation by flow cytometry. Additional exploratory objectives include the association of other clinical outcomes, baseline prognostic factors, and IWCLL-defined response with MRD status at multiple timepoints. Currently, enrollment is open. Clinical trial information: NCT04269902. [Table: see text]


Sign in / Sign up

Export Citation Format

Share Document