Meta-analysis of the adverse effects of trastuzumab in women with breast cancer

Lead Investigator: Alexandra Freeman, University of Cambridge
Title of Proposal Research: Meta-analysis of the adverse effects of trastuzumab in women with breast cancer
Vivli Data Request: 5869
Funding Source: None
Potential Conflicts of Interest: None

Summary of the Proposed Research:

The benefits of drugs for women with breast cancer are very well quantified, and are used to help decide which treatments might be of benefit to a particular woman. However, in order to make an informed choice, patients also need a similar quantification of the potential adverse effects of each drug. Currently there is no systematic review and meta-analysis of the results of drug trials in the major treatments for women with breast cancer (such as trastuzumab).

We are carrying out a series of analyses of data from trials that have been carried out using the drug trastuzumab (also known as Herceptin), which is used in women with breast cancer, to identify how many women suffer side effects and of what kind and severity.

To do this we are carrying out a number of analyses, one of which aims to identify whether there might be different side-effects in different kinds of women (for example, those who are pre- menopausal as opposed to post-menopausal) or whether there might be interactions between different drug regimes that could affect the side effects that women suffer (eg. do those taking certain chemotherapy drugs alongside trastuzumab suffer different side effects from those not taking them?).

For these two kinds of analysis, we need data for individual patients – not identifiable data, but data that tells us characteristics of each patient and what other drugs they were taking, alongside the data about the side effects that they suffered.

Statistical Analysis Plan:

review method is registered on PROSPERO (CRD42019146541)
Brief statistical methods:
1) For standard published adverse effects data from the long list of trials above, if the data allows, we propose a fixed-effects meta-analysis simply comparing the risks of suffering each adverse event in those taking trastuzumab therapy compared against those taking no trastuzumab therapy.

2) In a Bayesian network meta-analysis we will compare different models:
– Each different trastuzumab drug regime treated independently (including trastuzumab emtansine)
– Different delivery regimes (timings or dosages) considered the same, only different drug regimes (trastuzumab versus T-DM1 and concurrent with taxane or non-taxane chemotherapy) considered independently
– All trastuzumab-containing regimes pooled into one
For each adverse event, the different models will be compared to find the best-fitting model using all potentially relevant data available, using the Deviance Information Criterion. Adverse effects will be treated as significant if they show an odds ratio with a posterior median of 1.5 or greater and a lower 95% credible limit greater than 1 (the odds ratio being the ratio of the odds of the event between a treatment group and a control group receiving no chemotherapy).
To estimate the absolute risk of each event for those receiving trastuzumab, the odds ratios obtained from the meta-analyses will be applied to baseline absolute risks of the event obtained from trials in heathy women of the same demographic and from the control groups in the selected trials (women who were receiving other cancer treatments).
3) From individual patient-level data obtained for specific trials (i.e. the data here being requested), we will do subgroup analyses to investigate how the relative risks of adverse events associated with trastuzumab vary between :
– patients with different demographic or tumour characteristics (eg ethnicity, age, menopausal status, cancer stage)
– patients receiving different concurrent cancer therapies.
We will also look at clustering of side effects symptoms in order to be able to advise patients who suffer one symptom whether it is more likely that they will also suffer others.

Requested Studies:

A Randomized Three-Arm, Multicenter Comparison of 1 Year and 2 Years of Herceptin Versus No Herceptin in Women With HER2-Positive Primary Breast Cancer Who Have Completed Adjuvant Chemotherapy (HERA)
Sponsor: Roche
Study ID: NCT00045032
Sponsor ID: BO16348

Public Disclosures:

Jackson, C., Finikarides, L. and Freeman, A., 2022. The adverse effects of trastuzumab-containing regimes as a therapy in breast cancer: a piggy-back systematic review and meta-analysis. doi: 10.1371/journal.pone.0275321