Elsevier

European Journal of Cancer

Volume 50, Issue 16, November 2014, Pages 2771-2782
European Journal of Cancer

Original Research
Progression free survival rate at 9 and 18 weeks predict overall survival in patients with malignant pleural mesothelioma: An individual patient pooled analysis of 10 European Organisation for Research and Treatment of Cancer Lung Cancer Group studies and an independent study validation

https://doi.org/10.1016/j.ejca.2014.07.020Get rights and content

Abstract

Background

Response criteria have always been difficult to apply to malignant pleural mesothelioma (MPM), due to its unique pattern of growth. We developed some models to show that progression free survival rate (PFSR) could be a better predictor of overall survival (OS) than the response rate (RR) in MPM patients. The results were validated independently in the European Organisation for Research and Treatment of Cancer (EORTC) 08052, a phase II study in MPM.

Methods

Individual patient data from 10 EORTC-Lung Cancer Group (LCG) studies of first-line chemotherapy in MPM were pooled. Response to therapy was assessed according to World Health Organisation (WHO) criteria in all except the two most recent trials, which used Response Evaluation Criteria in Solid Tumours (RECIST). Landmark analyses (LA) at 9 weeks and 18 weeks after registration/randomisation were performed to assess the association between PFSR and OS. Independent validation of the results was conducted in EORTC 08052 study (82 patients) employing the same LA.

Results

All 10 studies (N = 523 patients) were included in the LA of PFSR at 9 and 18 weeks (PFSR-9 and PFSR-18). PFSR-9 and PFSR-18 were confirmed as predictors of OS, with hazard ratio (HR) of 0.37 (95% confidence interval (CI), 0.30–0.47) and 0.50 (0.38–0.65) and C-index of 0.62 and 0.58, respectively.

In the validation study, 28.4% achieved CR/PR and 77.8% had disease control (CR/PR/SD) as their best overall response. PFSR-9 and PFSR-18 weeks were both strongly correlated with OS (HR of 0.35 [80% CI, 0.25–0.49] and 0.46 (0.32–0.67) and C-index of 0.66 and 0.60, respectively).

Conclusion

PFSR-18 was strongly correlated and discriminated patients with better OS from the poorer prognosis patients. An earlier end-point, PFSR-9 was also strongly correlated to OS with better discriminating capacity. The results were independently validated.

Introduction

The occurrence of malignant pleural mesothelioma (MPM) parallels the exploitation and use of asbestos, with a mean latency period of approximately 40 years [1]. Thus, the incidence of MPM was expected to peak in the United States of America around 2010 [2], but continues to rise during the next 10 years in Europe [3]. Although the incidence of MPM should then decrease in developed countries, it will increase in developing countries, where the use of asbestos is poorly regulated, if at all [4]. Consequently, MPM will remain a major health problem worldwide for many decades.

While MPM was considered as refractory to treatments, two randomised phase III trials have demonstrated that first-line chemotherapy combining cisplatin and an antifolate (either pemetrexed [5] or raltitrexed [6]) significantly increased overall survival in patients with MPM compared to cisplatin alone. However, median overall survival does not exceed 12 months with these treatments and no standard of care is defined after failure of the latter. Research efforts must continue in the field of MPM to improve the efficacy of first-line therapy and identify effective second-line treatments. In phase II cancer clinical trials, the biological antitumour activity of drugs is generally assessed by measuring tumour shrinkage and the primary end-point is classically the response rate (response rate (RR), i.e. the proportion of complete and partial responses). However, this end-point has some disadvantages in the particular case of MPM. Due to its unique pattern of growth, response criteria have always been difficult to apply to MPM and for that reason a modified version of Response Evaluation Criteria in Solid Tumours (RECIST) has been developed [7]. Moreover, new drugs and notably targeted therapies are often more likely to work as cytostatic than cytotoxic agents. Thus, more relevant end-points should be considered in future MPM trials. From this perspective, one could propose to integrate disease stabilisation in the definition of drug activity.

The use of progression free/disease control as a primary end-point in phase II clinical trials has been previously explored. Lara et al. [8] showed that progression free rate was a stronger predictor of survival than tumour response rate in advanced non small cell lung cancer (NSCLC). Francart et al. [9] suggested that progression free rate at 18 weeks can be used as the primary end-point of phase II clinical trials in mesothelioma. In addition, Blayney et al. [10] explored the association of response with overall survival in patients with MPM.

We hypothesised that progression free survival rate (PFSR) could be a better predictor of survival than RR in MPM patients treated with first-line chemotherapy.

Section snippets

Patients and methods

The European Organisation for Research and Treatment of Cancer Lung Cancer group conducted between 1984 and 2001 a large screening programme to identify any active drug for first-line treatment of patients with MPM. A total of nine phase II trials were conducted. Later, one randomised phase III trial was implemented to compare a chemotherapy doublet to a monotherapy. Details concerning the chemotherapy regimens assessed in these 10 trials are summarised in Table 1.

The inclusion criteria were

Patient characteristics

Table 3 summarises the baseline characteristics of 523 patients from 10 EORTC studies where the model was developed and those of 82 patients from EORTC-08052 validation study. The majority of patients were male (83% versus 67%) with a median (range) age of 58 versus 55 (19–80 versus 22–77) years, had a performance status of 0/1 (86% versus 100%), epithelial histology (59%, versus 59%) and had stage III/IV disease (23/39% versus 23/43%) for all versus validation study, respectively. The baseline

Discussion and conclusions

This landmark survival analysis of 10 EORTC studies in patients with mesothelioma attempts to describe a reliable end-point which is simple to measure for early phase (phase II) clinical trials in mesothelioma, where traditional response measurement is challenging.

The data showed that PFSR status and response at 9 and 18 weeks strongly predict positive survival outcome. These findings hold up even after adjusting for known prognostic factors.

Although the models have been developed in

Conflict of interest statement

None declared.

Acknowledgement

This publication was supported by Fonds Cancer (FOCA) of Belgium.

References (15)

There are more references available in the full text version of this article.

Cited by (11)

  • EZH2 inhibitor tazemetostat in patients with relapsed or refractory, BAP1-inactivated malignant pleural mesothelioma: a multicentre, open-label, phase 2 study

    2022, The Lancet Oncology
    Citation Excerpt :

    Furthermore, the clinical and statistical teams involved in this study believed that overall response rate might not be adequately high enough to power the study. Thus, disease control rate was used as the primary endpoint, on the basis of meta-analyses of trials in patients with malignant pleural mesothelioma that found that disease control was strongly predictive of positive survival outcomes.7,8 The potential for long-term disease control to provide survival benefits independent of the overall response rate has also been observed in subsequent clinical trials.9

  • A multicentre randomised phase III trial comparing pembrolizumab versus single-agent chemotherapy for advanced pre-treated malignant pleural mesothelioma: the European Thoracic Oncology Platform (ETOP 9-15) PROMISE-meso trial

    2020, Annals of Oncology
    Citation Excerpt :

    Whilst mesothelioma progression may be difficult to reliably classify,14 we accounted for this potential bias by making BICR-PFS the primary end point, noting high concordance between BICR-assessed and investigator-assessed PFS. Moreover, others have argued that the PFS rate at 9 and 18 weeks is predictive of OS, giving an earlier end point readout, uncontaminated by uncontrolled post-progression therapies.15 Pembrolizumab crossover was accounted for by two well-defined methods, both demonstrating no significant OS benefit, although, we cannot fully exclude a small OS benefit that this methodology did not detect.

  • Effects of Reduction in Tumor Burden on Survival in Epithelioid Malignant Pleural Mesothelioma

    2018, Mayo Clinic Proceedings
    Citation Excerpt :

    Overall survival was significantly different among those with PR (12.8 months), SD (9.4 months), or PD (3.4 months) (P<.0001).16 A separate study based on many of the same patients assessed PFS rates at 9 and 18 weeks and found that both were predictive of OS.17 To limit selection bias, landmark analysis, removing patients who died or were censored before the landmark, was performed in these studies.

  • Malignant pleural mesothelioma: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up

    2015, Annals of Oncology
    Citation Excerpt :

    A study to evaluate ADI-PEG20, in combination with chemotherapy, in AS-negative mesothelioma [NCT02029690] is underway. Mutation of NF2 occurs in around 50% of mesotheliomas, sensitising inhibition of FAK [34–36]. Accordingly, this trial is stratifying patients by Merlin expression.Unlabelled Text

View all citing articles on Scopus

Presented as abstracts at ASCO Meetings in Chicago, June 2011 and June 2013.

View full text