scholarly journals Model-free approach to quantifying the proportion of treatment effect explained by a surrogate marker

Biometrika ◽  
2019 ◽  
Vol 107 (1) ◽  
pp. 107-122 ◽  
Author(s):  
Xuan Wang ◽  
Layla Parast ◽  
Lu Tian ◽  
Tianxi Cai
Keyword(s):  
Treatment Effect ◽  
Randomized Clinical Trials ◽  
Randomized Study ◽  
Model Misspecification ◽  
Asymptotic Properties ◽  
Surrogate Marker ◽  
Finite Samples ◽  
Model Free ◽  
Long Term Follow Up

Summary In randomized clinical trials, the primary outcome, $Y$, often requires long-term follow-up and/or is costly to measure. For such settings, it is desirable to use a surrogate marker, $S$, to infer the treatment effect on $Y$, $\Delta$. Identifying such an $S$ and quantifying the proportion of treatment effect on $Y$ explained by the effect on $S$ are thus of great importance. Most existing methods for quantifying the proportion of treatment effect are model based and may yield biased estimates under model misspecification. Recently proposed nonparametric methods require strong assumptions to ensure that the proportion of treatment effect is in the range $[0,1]$. Additionally, optimal use of $S$ to approximate $\Delta$ is especially important when $S$ relates to $Y$ nonlinearly. In this paper we identify an optimal transformation of $S$, $g_{\tiny {\rm{opt}}}(\cdot)$, such that the proportion of treatment effect explained can be inferred based on $g_{\tiny {\rm{opt}}}(S)$. In addition, we provide two novel model-free definitions of proportion of treatment effect explained and simple conditions for ensuring that it lies within $[0,1]$. We provide nonparametric estimation procedures and establish asymptotic properties of the proposed estimators. Simulation studies demonstrate that the proposed methods perform well in finite samples. We illustrate the proposed procedures using a randomized study of HIV patients.

scholarly journals Long-Term Follow-up of a Randomized Study of Oral Etoposide versus Viscum album Fermentatum Pini as Maintenance Therapy in Osteosarcoma Patients in Complete Surgical Remission after Second Relapse

Sarcoma ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-9 ◽  
Author(s):  
Alessandra Longhi ◽  
Marilena Cesari ◽  
Massimo Serra ◽  
Erminia Mariani
Keyword(s):  
Maintenance Therapy ◽  
Randomized Study ◽  
Survival Rates ◽  
Disease Free Survival ◽  
Viscum Album ◽  
Lymphocyte Subpopulation ◽  
Oral Etoposide ◽  
Long Term Follow Up

Background. In relapsed osteosarcoma, the 5-yr postrelapse disease-free survival (PRDFS) rate after the second relapse is <20%. In June 2007, a randomized study was started comparing oral etoposide vs Viscum album fermentatum Pini (an extract derived from the parasitic plant Viscum album L., European mistletoe) as maintenance therapy in patients with metastatic osteosarcoma in complete surgical remission after the second relapse. The primary endpoint was the PRDFS rate at 12 months (compared to the historical control rate). This is a long-term updated result. Patients and Methods. 10 patients received oral etoposide 50 mg/m2 daily for 21 days every 28 days for 6 months, and 9 patients received Viscum album fermentatum Pini 3 times/wk subcutaneously for 1 year. The study closed early in July 2011 due to insufficient recruitment. Lymphocyte subpopulations were analyzed at T0, T3, T6, T9, and T12 months. Results. On 30 June 2019, at a median follow-up ITT of 83 months (range 3–144 ms), a median PRDFS of 106 ms (2–144) was observed in the Viscum arm with 5/9 patients who never relapse vs a PRDFS of 7 months (3–134) in the etoposide arm (all patients in the Etoposide arm relapsed) (hazard ratio HR = 0.287, 95% CI: 0.076–0.884, p=0.03). Model forecast 10-yr overall survival rates as 64% in the Viscum arm and 33% in the etoposide arm. Lymphocyte subpopulation counts (CD3, CD4, and CD56) showed an increase in the Viscum arm while a decrease was observed in the etoposide arm during treatment. Conclusions. After 12 years from the start of the trial, the patients in the Viscum arm continue to show a considerably longer PRDFS compared to oral etoposide, and a trend for an advantage in OS is evident even if the number of treated patients is too small to draw conclusions. Viscum as maintenance treatment after complete surgical remission in relapsed osteosarcoma should be further investigated and compared with other drugs.

scholarly journals Continued Excellent Outcomes in Previously Untreated Patients With Follicular Lymphoma After Treatment With CHOP Plus Rituximab or CHOP Plus 131I-Tositumomab: Long-Term Follow-Up of Phase III Randomized Study SWOG-S0016

2018 ◽  
Vol 36 (7) ◽  
pp. 697-703 ◽  
Author(s):  
Mazyar Shadman ◽  
Hongli Li ◽  
Lisa Rimsza ◽  
John P. Leonard ◽  
Mark S. Kaminski ◽  
...  
Keyword(s):  
Overall Survival ◽  
Acute Myeloid Leukemia ◽  
Myeloid Leukemia ◽  
Randomized Study ◽  
Phase Iii ◽  
Second Malignancies ◽  
Long Term Follow Up ◽  
Acute Myeloid

Purpose SWOG S0016 was a phase III randomized study that compared the safety and efficacy of R-CHOP (rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone) with CHOP-RIT (CHOP followed by consolidation with iodine-133–tositumomab radioimmunotherapy) for previously untreated patients with follicular lymphoma. Understanding the long-term outcome of patients provides a benchmark for novel treatment regimens for FL. Patients and Methods Between 2001 and 2008, 531 previously untreated patients with FL were randomly assigned to receive either six cycles of R-CHOP or six cycles of CHOP-RIT. Patients with advanced-stage disease (bulky stage II, III, or IV) of any pathologic grade (1, 2, or 3) were eligible. Results After a median follow-up of 10.3 years, 10-year estimates of progression-free and overall survival were 49% and 78% among all patients, respectively. Patients in the CHOP-RIT arm had significantly better 10-year progression-free survival compared with patients in the R-CHOP arm (56% v 42%; P = .01), but 10-year overall survival was not different between the two arms (75% v 81%; P = .13). There was no significant difference between the CHOP-RIT and R-CHOP arms in regard to incidence of second malignancies (15.1% v 16.1%; P = .81) or myelodysplastic syndrome or acute myeloid leukemia (4.9% v 1.8%; P = .058). The estimated 10-year cumulative incidences of death resulting from second malignancies were not different (7.1% v 3.2%; P = .16), but cumulative incidence of death resulting from myelodysplastic syndrome or acute myeloid leukemia was higher in the CHOP-RIT arm compared with the R-CHOP arm (4% v 0.9%; P = .02). Conclusion Given these outstanding outcomes, immunochemotherapy should remain the standard induction approach for patients with high-risk FL until long-term follow-up of alternative approaches demonstrates superiority.

scholarly journals LTBK-12. EORTC 26951, RANDOMIZED STUDY OF ADJUVANT PCV AFTER 59.4 GY RADIOTHERAPY: VERY LONG TERM FOLLOW-UP

2019 ◽  
Vol 21 (Supplement_6) ◽  
pp. vi285-vi285
Author(s):  
Martin van den Bent ◽  
Khe Hoang-Xuan ◽  
Alba Brandes ◽  
Johan Kros ◽  
M C M Kouwenhoven ◽  
...  
Keyword(s):  
Randomized Study ◽  
Progression Free Survival ◽  
Phase Iii ◽  
Term Survival ◽  
Long Term Survival ◽  
Kaplan Meier ◽  
Tumor Group ◽  
Long Term Follow Up

Abstract BACKGROUND Between 1995 and 2002 the EORTC Brain Tumor Group conducted a prospective phase III study on adjuvant procarbazine, CCNU and vincristine (PCV) chemotherapy in anaplastic oligodendroglioma (AOD). A mature follow-up presented in 2012 showed survival benefit of the addition of PCV, in particular in 1p/19q co-deleted tumors and tumors with MGMT promoter methylation. We now present very long term follow-up. MATERIALS AND METHODS Patients were eligible if locally diagnosed with a newly diagnosed AOD. They were randomized between radiotherapy (RT, 33 x 1.8 Gy) and the same RT followed by 6 cycles PCV (RT/PCV). Primary endpoints were overall survival (OS) and progression free survival (PFS). 1p/19q status (FISH) was determined in 300 patient. Kaplan- Meier technique and Cox modeling were used for long term survival analysis. Primary analyses were adjusted for known prognostic factors. For other analyses no adjustment was performed. RESULTS With 368 patients included, a median follow-up of 18.4 years and 307 (83%) survival events, median and 20-year survival after RT/PCV versus RT alone were 42.3 mo and 16.8% vs 30.6 months and 10.1% (HR 0.78; 95% CI (0.63, 0.98), adjusted p=0.06). Eighty patients were 1p/19q codel of which 26 (33%) were still alive, in this subgroup median and 20-year survival after RT/PCV versus RT alone were 14 years and 37.1% versus 9.3 years and 13.6% (HR 0.60, 95% CI (0.35, 1.03), unadjusted p=0.06). Twenty year PFS in 1p/19q codel was 31.3% in RT/PCV treated patients and 10.8% in RT only treated patients (HR 0.49, 95% CI (0.29, 0.83), unadjusted p=0.007). In the 1p/19q codel subgroup age, WHO PS and necrosis at pathology were identified to be of independent prognostic value for OS. CONCLUSION This long term analysis confirms the earlier conclusions and provides data on long term survival in this patient group. In 1p/19q codel patients treated with RT/PCV, the 20-year PFS and OS rates are 31% and 37% respectively.

Sign in / Sign up

Export Citation Format

Share Document