ESMO 2017: Clinical Trial Toxicity Reporting by Investigators May Not Reflect Patients’ Viewpoint

Patient experience according to HRQoL questionnaire does not mirror investigator toxicity reporting for many symptoms

An analysis comparing health related quality of life (HRQoL) by patient assessment and toxicity reporting by physicians in clinical trials demonstrated little agreement on many symptoms between the two, indicating that patients and investigators may not be on the same page regarding the overall patient clinical trial experience.

Findings from an analysis comparing patient-reported outcomes and investigator assessed toxicities in a large phase III adjuvant breast cancer trial were presented at ESMO 2017, the Annual Congress of the European Society for Medical Oncology in Madrid, Spain.

Yvonne Brandberg of the Oncology-Pathology Department, Karolinska Institutet in Stockholm, Sweden, and a team of investigators reviewed the agreement between toxicity items reported by physicians using the Common Terminology Criteria for Adverse Events, version 3.0. (CTCAEs) and patient-reported HRQoL items in the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30).

Data collected from assessment made at the end of treatment in PANTHER, a randomised phase III trial, was used for this analysis. The trial evaluated dose dense versus standard administration of adjuvant chemotherapy in women with high-risk breast cancer, the majority of whom had node positive disease.1

For the comparisons made in this analysis, three researchers individually selected items from the EORTC QLQ-C30 they considered to be comparable to CTCAE criteria. The relationship was based on ordinal data that were analysed by Goodman and Kruskal gamma methods.

The analysis included data from 1428 event-free patients and investigated the relation between 13 toxicities and 36 EORTC QLQ-C30 items, with some patients having more than one toxicity.

Strong and moderate agreement seen for diarrhoea, vomiting and fatigue only

The investigators found a strong relation (0.5 to 1.0) or strong agreement between physician assessed toxicity and patient-reported outcomes for the HRQoL scale questions 17 “Have you had diarrhoea?” and 15, “Have you vomited?”. The relation between the reports for diarrhoea was 0.53 (range 0.47 to 0.59) and the relation for vomiting was 0.50 (range 0.39 to 0.62).

The relation was moderate (0.30-0.49) for HRQoL questions 10, 12, and 8 relating to fatigue (relation 0.35, 0.35, and 0.42, respectively).

Physician rated “pain” was moderately related to patient ratings on question 9, the relation was 0.31 (0.25 to 0.37).

All of the comparisons between the nine remaining physician-reported toxicities and patient-reported HRQoL scores and toxicity showed weak or no relation.


The investigators found little agreement between physician evaluated toxicities by CTCAE criteria and the true patient experience, as reported on HRQoL questions, prompting them to conclude that these findings should raise concerns on how to best evaluate HRQoL and toxicities in clinical trials.

Karin Jordan of the University of Heidelberg, Germany who discussed the study results thanked the authors for bringing in a large trial a valuable perspective that can inform understanding of treatment toxicities. For her the questions arising from this study are who is right: CRO or PROM, patient or clinician, and are PROMs really that helfpul. Patient and clinician reports of symptoms and particularly symptomatic toxicities during cancer treatment provide discrepant yet complementary data. There are two potential views of the situation: 1. more patient-centered view: the patient is always right by definition because nobody (not even the most sensitive clinician) can truly know another person’s subjective experience; 2. more traditional view: the clinicians should be considered right because they have an “objective” perspective based on experience and training. She questioned if combination of these views are possible. Three potential approaches are: 1) “Independent reporting,” in which patient and clinician toxicity data are collected, analyzed, and reported completely separately from each other; 2) “Merged reporting,” in which patient and clinician data are collected separately and then merged analytically into a single metric; and 3) “Collaborative reporting,” in which patients directly report symptomatic toxicity information, which is then provided to clinicians to inform their CTCAE reporting.


The study was funded by the Swedish Cancer Society.


  1. Foukakis T, et al. JAMA 2016;316(18):1888-1896.


1422O – Brandberg Y, et al. CTCA toxicity scoring and EORTC Quality of Life Questionnaire – a comparison of physicians’ and patients’ scoring of toxicity in the "Panther trial".