scholarly journals The impact of ignoring Interval censoring in progression-free survival in cancer trials: a systematic review

Author(s):  
Xiawen Zhang ◽  
Eleanor Pullenayegum ◽  
Kelvin Kar-Wing Chan

Introduction & Objective: From statistical literature, the bias in treatment effect from ignoring interval censoring in Progression-free survival (PFS) is demonstrated. However, the impact on estimators caused by interval censoring is not carefully took account and investigated by researchers in practice. The objective of this study is to examine the impact of accounting for interval censoring in practice among RCTs used to support FDA approvals anti-cancer drugs between the years 2005 and 2019 that used PFS as an endpoint. Methods: In this systematic review, the differences of hazard ratios between two methods: considering and ignoring interval censoring, are visualized by Kaplan-Meier survival curves and estimated from a Cox proportional hazard model of 87 RCTs. With assumption that these differences and mean differences (bias) follow a normal distribution, limits of agreement of differences and confidence interval of bias are used to represent agreement of two methods. Results: Limits of agreement of difference range from -0.044 to 0.0615, while confidence intervals for the bias range from 0.0026 to 0.0145, which does not include zero, resulting in estimated treatment effect differs for two methods. Conclusion: In general, bias caused by interval censoring in treatment effect exists with large sample studies. Focusing on individual clinical trials, limits of agreement can provide more information for researchers to make decision on how to account for interval censoring.

2019 ◽  
Vol 37 (15_suppl) ◽  
pp. 9105-9105 ◽  
Author(s):  
Michael E. Menefee ◽  
Yutao Gong ◽  
Pallavi Shruti Mishra-Kalyani ◽  
Rajeshwari Sridhara ◽  
Bindu Kanapuru ◽  
...  

9105 Background: Docetaxel is a common comparator arm to test novel therapies in post-platinum mNSCLC trials. The advent of Real World Evidence (RWE) has renewed interest in the use of synthetic control arms (control arms from previously conducted randomized trials) to improve accrual to trials and increase patient access of promising experimental agents. We reviewed legacy second-line (2L) mNSCLC trials to assess the impact of switching docetaxel control arms from one trial to another and compare to an experimental regimen. Methods: We identified 5 contemporary 2L trials that enrolled 2013 patients receiving an experimental therapy vs. docetaxel: 5 immunoncology head-to-head trials (one with 2 arms) and one anti-VEGF add-on trial. Kaplan-Meier curves of overall survival (OS) and progression-free survival (PFS) were produced for docetaxel controls. We calculated OS and PFS hazard ratios and 95% confidence intervals for each synthetic trial. A pooled doc arm was also compared with each experimental agent. Results: See Table. Conclusions: Both individual and pooled docetaxel switching of control arms approximated the original OS HR and 95% CI. Methods such as bootstrapped sampling and propensity score matching will be performed in an effort to more closely approximate the original trial characteristics. [Table: see text]


2013 ◽  
Vol 31 (15_suppl) ◽  
pp. e13034-e13034
Author(s):  
Menal Bhandari ◽  
Ajeet K Gandhi ◽  
Pramod Kumar Julka ◽  
Chitra Sarkar ◽  
Dayanand Sharma ◽  
...  

e13034 Background: This study assesses the impact of 6 cycles of adjuvant TMZ (conventional arm) versus 12 cycles (Extended arm) on Progression free survival (PFS), evaluate the toxicity and correlate the outcome with EGFR, P53 and MIB I labelling Index. Methods: Between December 2010 to October 2012, 36 post operative patients of Glioblastoma between age 18-65 years and Karnofsky Performance Score (KPS) ≥ 70 were included. Patients were randomized to receive Radiation with a dose of 60 Gray in 30 fractions over 6 weeks at 2 gray/fraction with concomitant TMZ (75 mg/m2/day) and Adjuvant therapy with either 6 or 12 cycles of TMZ(150 mg/m2 for 5 days, 28 days cycle). Patients were then assessed monthly clinically and imaged with MRI/CT every 3 monthly or when symptomatic. Toxicity was assessed using CTCAE version 3.0. Statistical Analysis was done using SPSS version 17.0.Kaplan Meier method was used for analysis of survival and log rank test was used for assessing the impact of variables on survival. Results: Of 36 patients, 18 patients were treated in each arm. Median age and KPS in both the arms was 47 years and 80 respectively. 44 % patients in the conventional arm and 50% patients in the Extended arm underwent complete surgical resection. 22% patients in the conventional arm and 28% in the extended arm did not complete their intended treatment. Grade ¾ Thrombocytopenia was seen in 16% in the extended arm and 0% in the conventional arm.EGFR, P 53 and MIB 1 >20% was seen in 26%, 45% and 20% patients respectively, overall. Median follow up was 18 months for both the arms (Range 10-23 months).At last follow up,8 patients in each arm had progression. Median PFS was 10 months vs.18.4 months (p 0.47) in conventional and extended arm respectively. On Univariate analysis, patients with KPS ≤ 80 had poorer survival than those >80 (Median PFS 9.5 Months vs. 16.9 Months; p 0.02).Age, extent of resection, EGFR, P53, MIB 1 did not significantly alter survival in the two treatment groups. Conclusions: Our study showed that schedule of extended Temozolomide is well tolerated by patients and tend to have better progression free survival. Further prospective randomized studies are needed to validate the findings of our study.


2013 ◽  
Vol 31 (15_suppl) ◽  
pp. e19022-e19022
Author(s):  
Liza Cosca Villaruz ◽  
Mark A. Socinski ◽  
Jyoti D. Patel ◽  
Larry Leon ◽  
Sebastien Hazard ◽  
...  

e19022 Background: Progression-free survival (PFS) is a key trial end point for clinical practice, as it relates to a change of treatment line. Grade 4 progression-free survival (G4PFS; defined as time from treatment start to the earlier of progressive disease [PD], onset of a G4 adverse event [AE], or death from any cause) is a composite end point incorporating a measure of tolerability to PFS. This post hoc analysis evaluates G4PFS and the effect of G4 AEs on PFS and overall survival (OS) in patients (pts) enrolled in PointBreak (PB) and AVAiL. Methods: Pts in PB were randomized to bevacizumab (BEV) 15 mg/kg q3w with carboplatin and paclitaxel (CP) or pemetrexed (CPem) for ≤4 cycles. Eligible pts received either BEV or BEV + Pem q3w until PD or unacceptable toxicity. Pts in AVAiL received cisplatin and gemcitabine for ≤6 cycles and either placebo or BEV (7.5 or 15 mg/kg) q3w until PD. AEs were graded via NCI-CTCAE v3.0. Kaplan-Meier and Cox model methods were used to estimate medians and hazard ratios (HRs) for PFS, G4PFS, and OS. PFS and OS were also compared in each arm for pts with occurrence of a G4 AE before week 12 of treatment vs those without. Results: Of those receiving BEV, ~30% of AVAiL pts and 38% of PB pts had a G4 AE. Uncomplicated neutropenia was the most common G4 AE in the CP + BEV arm of PB (26%) and in the BEV arms of AVAiL (11%). PFS, G4PFS, and outcomes by G4 AE occurrence are shown (Table). Conclusions: In both the PB and AVAiL trials, median G4PFS was numerically shorter than median PFS. G4 AE occurrence, however, did not affect subsequent PFS or OS in either trial. Clinical trial information: NCT00762034 and NCT00806923. [Table: see text]


Blood ◽  
2018 ◽  
Vol 132 (Supplement 1) ◽  
pp. 4267-4267
Author(s):  
Pongtep Viboonjuntra ◽  
Arnuparp Lekhakula ◽  
Kanchana Chansung ◽  
Chittima Sirijerachai ◽  
Pimjai Niparuck ◽  
...  

Abstract Introduction : To date, the ELN recommendation and NCCN guidelines are the principle mile stones to follow up the treatment response and to make the decision of TKIs switching. However, in real life practice, many factors influence changing the real switching date from the date had an indication. This study aims to analyze the impact of early switching to second line TKI, nilotinib, in real life practice, for the CML patients who failed, had sub-optimal response or were intolerant to imatinib. Methods : This prospective study was conducted through 7 medical centers in Thailand between 1st of September 2009 and 31st of August 2011. Adult CML patients of age ≥ 18 years old, in chronic and accelerated phase, who had failure, suboptimal response or intolerance to imatinib, based on ELN 2009 guideline, were included and were eligible with nilotinib 400 mg twice daily. Prospective data collection for 24 months of each patient was performed. The main objective was to identify the impact of early switching to nilotinib on major molecular response (MMR). The other objectives were to observe the efficacy of nilotinib including overall survival, progression free survival and the safety. The survival results were presented as Kaplan-Meier survival curves. For the comparison of the treatment groups, the Kaplan-Meier estimator with the corresponding log-rank test for equality of survivor functions across treatment group was applied. Results : The final 108 cases were analysed. The median age was 47 (17-79) years with the proportion of male to female of 1.4:1 respectively. The median duration of the prior imatinib treatment was 18 months (2-142 months). The median duration between the date of indication and the date of real switching was 3.1 months (0-62.8 months) with 50% changing less than 3 months, 26.9% between 3 months and 12 months, and 23.1% changing longer than 12 months. The indication of switching included 63.6% failure to imatinib, 29% intolerance to imatinib and 7.4% suboptimal to imatinib. On the nilotinib switching, 70.4% completed 24 months follow-up, and 29.6% discontinued treatment mostly because of unsatisfactory results or adverse events. Evaluation was made every 3 months based on 2009 ELN recommendation. At 3 months, 57%, 20%, and 8% of the patients achieved CHR, CCyR and MMR, respectively. Those who did not achieve CHR at 3 months never achieved MMR, while 86 % of those who achieved CCyR at 3 months achieved MMR. All CML achieving MMR at 3 months had sustained MMR throughout the study period (24 months). Imatinib suboptimal response had better outcome than imatinib failure and imatinib intolerance groups. A preliminary analysis of BCR-ABL mutation was performed on 90 cases, and mutations were found on 21 cases. Two of them were T315I which were excluded from the study. The cases with mutation had poorer response to treatment than those without mutation. There was one case with initial G250E mutation developing T315I mutation after treatment with nilotinib. At 24 months, one case progressed to accelerated phase and 3 cases progressed to blastic transformation. The 2-year overall survival and 2-year progression-free survival and were 98.9% and 96.9% (figure 1 and 2), respectively. The interquatile analysis was done to identify the groups of cumulative MMR according to the duration between the date of indication and the date of real switching to nilotinib. The patients who switched to nilotinib within 12 months after date of indication could have a greater chance to achieved MMR than those who switched treatment later than 12 months (p(log-rank) = 0.002) (figure 3). Skin rash, musculoskeletal pain, and infection were the three most common non-hematologic adverse events, However, most of them were grade 1-2, except for 4 cases with grade 3-4 infections. Grade 3-4 hematologic adverse events included thrombocytopenia (12%), neutropenia (11%), anemia (5%) and leucopenia (4%), and most of them were manageable. Although biochemical abnormalities were commonly found, most of them were mild. Conclusions : Nilotinib, as a second line treatment showed excellent efficacy and tolerability. Indication for nilotinib treatment, initial mutation status and depth of response at 3 months after treatment can predict outcomes of the patients. However, the patients will have a greater chance to achieve MMR if they switched to nilotinib within 12 months after the date of indication for changing. Disclosures No relevant conflicts of interest to declare.


2019 ◽  
Vol 37 (15_suppl) ◽  
pp. 8565-8565
Author(s):  
Giulia Pasquini ◽  
Antonio Chella ◽  
Claudia Menichelli ◽  
Marco Lucchi ◽  
Maria Grazia Fabrini ◽  
...  

8565 Background: Thymomas but not thymic carcinomas can benefit of cytoreductive surgery even if a complete resection is not achievable. Surgical resection of pleural metastases, the most common site of progression, can be performed in selected patients. We evaluated the outcome of stereotactic body radiation therapy (SBRT) for treatment of pleural metastases in patients’ not eligible for surgery. Methods: We retrospectively identified 22 patients treated with SBRT for pleural metastases between 2004 and 2019. According to RECIST criteria, time to local failure and progression free survival (PFS) were calculated using Kaplan-Meier estimation. Results: Twelve of the 22 patients were male. The median age was 40 years (range 29-73). There were 1 A, 3 AB, 3 B1, 3 B2, 3 B2/B3 and 9 B3 thymomas. The Masaoka stage at diagnosis was IIA in 2, IIB in 7, III in 5, IVA in 7 and IVB in 1 patient. Pleural metastases and primary tumor were synchronous in 8 patients. Thymectomy was performed in 21 patients. Seven patients received pre-operative chemotherapy and 12 post-operative radiotherapy. One patient received chemotherapy and radiotherapy after a macroscopically incomplete thymectomy. Five patients had a single pleural metastatic site and 17 presented multiple localizations. Sixteen patients received SBRT on multiple sites of pleural metastases. At the time of the analysis a patient received SBRT exclusively on one of 3 pleural metastases. The median dose of radiation was 30Gy (range 25-40) given in 3 fractions. Ten patients experienced a progression of treated lesions with a median time to local failure of 25.5 months (95%CI 20.9-30.1). The median PFS was 20.4 months (95%CI 10.7-30). There were not significant differences in PFS between patients diagnosed with synchronous and metachronous metastases (p=0.477), across those treated with chemotherapy or naive (p=0.189) and between those who received or not a previous surgical resection of the pleural metastases (p=0.871). Conclusions: SBRT of pleural metastases is feasible and offer an interesting local control of diseases. The impact of this treatment on patients’ survival is hardly predictable because of the heterogenous clinical behavior of thymomas.


2014 ◽  
Vol 32 (27) ◽  
pp. 3068-3074 ◽  
Author(s):  
Federico Campigotto ◽  
Edie Weller

Informative censoring in a progression-free survival (PFS) analysis arises when patients are censored for initiation of an effective anticancer treatment before the protocol-defined progression, and these patients are at a different risk for treatment failure than those who continue on therapy. This may cause bias in the estimated PFS when using the Kaplan-Meier method for analysis. Although there are several articles that discuss this issue from a theoretical perspective or in randomized phase III studies, there are little data to demonstrate the magnitude of the bias on the estimated quantities from a phase II trial. This article describes the issues by using two oncology phase II trials as examples, evaluates the impact of the bias using simulations, and provides recommendations. The two trials were selected because they demonstrate two different reasons for censoring. Simulations show that the magnitude of the bias depends primarily on the proportion of patients who are informatively censored and secondarily on the hazard ratio between the group of patients who remain on study and the group of patients who are informatively censored. Recommendations include using an alternative end point, which includes inadequate response and initial signs of clinical progression as treatment failure, and a competing risk analysis for studies in which competing events preclude or modify the probability of observing the primary event of interest. If informative censoring cannot be avoided, then all patients should be observed until progression, and sensitivity analyses should be used as appropriate.


2020 ◽  
Author(s):  
Jigao Feng ◽  
Yibin Ouyang ◽  
Dedong Xu ◽  
Qinglong He ◽  
Dayuan Liu ◽  
...  

Abstract Background: lncRNA MIR17HG was upregulated in glioma, and involved glioma proliferation, migration, invasion and promoted apoptosis. However, the role of MIR17HG polymorphisms on the occurrence and prognosis of glioma is not obvious. Methods: In the study, 592 glioma patients and 502 control subjects conducted. Agena MassARRAY platform was used to detect the genotype of MIR17HG polymorphisms. Logistic regression analysis was used to evaluate the relation of MIR17HG SNPs to glioma risk by odds ratio (OR) and 95% confidence intervals (CIs). Kaplan–Meier curves, Cox hazards models were performed for assessing the role of these SNPs in glioma prognosis by hazard ratios (HR) and 95% CIs. Results: We found that rs7318578 (OR = 2.25, p = 3.18x10-5) was significantly associated with glioma susceptibility. Rs17735387 (OR = 1.53, p = 9.05x10-3) and rs7336610 (OR = 1.35, p = 0.016) had a higher glioma susceptibility in the subgroup with age < 40 years. More importantly, rs17735387 (HR = 0.82, log-rank p = 0.026) improved glioma prognosis. GA genotype of rs17735387 had a better overall survival (HR = 0.75, log-rank p = 0.013) and progression free survival (HR = 0.73, log-rank p = 0.032) in patients with Ⅰ-Ⅱ glioma. Conclusion: Our study firstly reported that MIR17HG polymorphisms, especially rs7318578, might be risk factors for glioma susceptibility and rs17735387 improved the prognosis of glioma among Chinese Han population, which might help to enhance the understanding of MIR17HG gene in gliomagenesis.


Blood ◽  
2005 ◽  
Vol 106 (11) ◽  
pp. 1159-1159 ◽  
Author(s):  
Muzaffar H. Qazilbash ◽  
Marilyn S. Davis ◽  
Aleman Ana ◽  
Linda Roden ◽  
Floralyn Libunao ◽  
...  

Abstract Backround: Arsenic trioxide (ATO), an active agent against multiple myeloma, has been shown to be synergistic with melphalan both in vitro and in vivo. We conducted a phase I/II trial to determine the safety and efficacy of a combination of arsenic trioxide, melphalan and ascorbic acid (AA) as prepaprative regimen in patients undergoing high-dose therapy (HDT) and autologous hematopoietic progenitor cell transplantation (AHPCT) for multiple myeloma (MM). We also assessed the impact ATO levels on melphalan pharmacokinetics (PK), engraftment and toxicity. Methods: Twenty-five patients with secretory myeloma (11 females, 14 males median age: 53, range: 49 – 69) were treated b/w 4/04 and 1/05. All patient received melphalan 100 mg/m2 IV on days -4 and -3 and AA 1000 mg/day IV on days −9 to −3. Patients were randomized to 3 arms; no ATO (arm 1), ATO 0.15 mg/kg IV on days −9 to −3 (arm 2) and ATO 0.25 mg/kg IV on days −9 to −3 (arm 3). Seven patients had a prior autograft. Median CD34 cells dose infused was 4.4 x 106/kg (range 2.3–10.9). Results: Patients were evenly matched except for a high median β2m level (3.6 vs. 2.4 in arms 1 and 2, p=0.04) in arm 3. With a median F/U of 7.1 months post autograft, no dose-limiting toxicity or non-relapse mortality was seen. Median ATO level on day 0 in arms 1, 2 and 3 were 0.2, 26.3 and 46.2 ng/ml, respectively. Toxicity was limited grade I or II nausea, vomitting and diarrhea. Median time to neutrophil engraftment (ANC &gt;500/dl) was 9 days. There were no engraftment failures or delays in the ATO arms. Response rates (RR) are shown in Table 1. With a median F/U of 7.1 months (range, 6.4 – 8.9 months), the progression-free survival (PFS) and overall survival (OS) are as shown in Figure 1. There was no significant difference in RR, PFS or OS between the 3 arms. Melphalan PK was not altered by ATO pretreatment. Conclusions: Arsenic trioxide, given in combination with melphalan and ascorbic acid as preparative regimen, is safe and well tolerated. A longer follow up is needed to determine the impact of this combination on survival. Response Rate at 3-Month Evaluation Response at 3 months CR PR MR SD PD p = 0.55 CR = complete response, PR = partial response, MR = minimal response, SD = stable disease, PD = progressive disease Arm 1 (no ATO) 1 5 0 1 1 Arm 2 (ATO 0.15) 1 5 2 0 1 Arm 3 (ATO 0.25) 0 7 0 0 0 Figure 1. The Kaplan-Meier estimates for progression-free survival probability (N=25). Figure 1. The Kaplan-Meier estimates for progression-free survival probability (N=25). Figure 2. The Kaplan-Meier estimates for overall survival probability (N=25). Figure 2. The Kaplan-Meier estimates for overall survival probability (N=25).


2016 ◽  
Vol 34 (4_suppl) ◽  
pp. 770-770 ◽  
Author(s):  
May Thet Cho ◽  
Jonathan Kessler ◽  
John Park ◽  
Gagandeep Singh ◽  
Yi-Jen Chen ◽  
...  

770 Background: The use of selective internal radiation therapy (SIRT) with SIR-Spheres in chemotherapy-resistant colorectal cancer liver metastases (CRC-L) has been associated with favorable progression free survival (PFS) and overall survival (OS) when given alone or concurrently with chemotherapy. However, no prospective studies exist for concurrent SIRT and chemotherapy (SIRT-CT) vs. SIRT alone. We conducted a single institute retrospective trial to compare the effect of SIRT-CT to SIRT alone on liver PFS in patients with CRC-L. Methods: Patients (pts) with CRC-L treated with SIR-Spheres at City of Hope between 2009 and 2014 were identified. CRL-L patients treated with SIRT-CT or with SIRT were excluded if they received, following radioembolization, any chemotherapy/targeted regimen on which they did not previously progress. This strategy was adopted to minimize the impact of post-SIRT systemic therapy bias on SIRT-CT/SIRT liver PFS outcome. Pts characteristics included demographics, liver involvement pattern, and lines of prior chemotherapy. Liver PFS, response rate, and toxicities were compared between SIRT-CT and SIRT arms. Kaplan-Meier estimation was used for PFS analysis. Results: 48 CRC-L pts were treated with SIR-spheres; 27 satisfied the post-SIR-spheres systemic treatment criteria (14 SIRT-CT, 13 SIRT) and were included on this study. Pts characteristics included: age (median = 62; range 52-80), sex (18 males), primary site (colon: 23), hepatic disease burden (23 bilobar), 5-FU resistance/intolerance (25/2), and extrahepatic disease (22). Pts characteristics were similar between treatment arms, except for median prior therapies (SIRT-CT = 3, SIRT = 2). No SIRT-CT or SIRT associated ≥ grade 3 toxicities were noted. Disease control rates were 84% (2/13 PR; 9/13 SD) and 14% (2/14 SD on the SIRT-CT and SIRT arms, respectively (p = 0.001). Median PFS in the liver was 176 days in the SIRT-CT group vs. 91 days in the SIRT group (p = 0.0006). Conclusions: In patients with 5-FU refractory CRC-L, SIRT-CT is associated with an increased disease control rate and a prolonged DFS in comparison with SIRT alone.


Sign in / Sign up

Export Citation Format

Share Document