ToxT Analysis of Real-Life Data Improves Toxicity Characterisation

Toxicity over Time methodology could improve characterisation and management of side effects of oncology agents

medwireNews: Study findings showing the application of the Toxicity over Time (ToxT) approach in two oncology clinical trials suggest the method may offer a more useful picture of side effects than traditional adverse event profiles.

ToxT uses longitudinal techniques, including graphs and tables, to visualise the impact of side effects over the course of treatment, revealing information such as first occurrence and worst grade of toxicity, and the presence of persistent low-grade adverse effects, the researchers explain in The Lancet Oncology.

“ToxT has a role in the clinic, for optimally counselling individual patients on the anticipated side-effects of a given treatment, and in clinical trials, to better depict adverse events of novel agents or combinations, and to make trials more patient-centred”, write Gita Thanarajasingam, from the Mayo Clinic in Rochester, Minnesota, USA, and co-authors.

“This type of information might also be central to the process of securing regulatory approval for novel agents in the future.”

The team applied ToxT analysis to adverse event information collated from a trial assessing oxaliplatin , 5-fluorouracil (5-FU) and irinotecan combinations in patients with metastatic colorectal cancer, and a second study comparing venlafaxine and placebo for hot flush control in breast cancer survivors.

ToxT revealed that patients given irinotecan plus oxaliplatin were significantly more likely to experience early nausea than those given 5-FU and oxaliplatin, with an average nausea grade of 1.1 versus 0.6 in the first cycle.

And analysis of event charts demonstrated that patients experienced grade 2 or more severe diarrhoea earlier with irinotecan plus oxaliplatin than with 5-FU plus oxaliplatin, with 26.2% versus 8.2% affected after 1 month; area under the curve analysis confirmed that mean diarrhoea grade was higher with irinotecan plus oxaliplatin throughout treatment.

Indeed, the butterfly plot comparing average grade of nausea, vomiting, diarrhoea, alopecia and neutropenia during treatment “makes clear visually that gastrointestinal toxic effects in particular were substantially higher grade in patients on [irinotecan plus oxaliplatin] than those on [5-FU plus oxaliplatin], particularly in early treatment cycles”, the authors write.

Similarly, although breast cancer survivors who received venlafaxine had a comparable risk of dry mouth to that of placebo-treated controls at the start of treatment, affecting 13% and 22% at week 1, venlafaxine was associated with a significantly higher rate of the side effect by week 5, at 49% versus 2%.

The author of an accompanying comment notes that while some of the methods used in ToxT are readily available, the findings are “important”, citing the clarity of their graphical displays to convey complex data and their “brilliant” illustration of the ToxT package using real-world data.

Tito Mendoza, at the University of Texas MD Anderson Cancer Center in Houston, USA, encourages the application of the ToxT package to the US National Cancer Institute Common Terminology Criteria for Adverse Events and its complementary Patient Reported Outcomes version.

He highlights that the grading system does not fully characterise treatment toxicity, writing that “it might seem acceptable for 10% of patients participating in a 100 day trial of a particular treatment to experience grade 3 diarrhoea (eg, increase of ≥7 stools per day over baseline), but not acceptable if 20% of the patients experienced less-severe grade 2 diarrhoea (eg, increase of 4–6 stools per day over baseline) 50 or more times withinthe same timeframe.”

Tito Mendoza concludes: “Having complete information about a treatment’s adverse effects is not only helpful for patients, but also crucial to the treating physician’s ability to offer optimum symptom management.”


Thanarajasingam G, Atherton PJ, Novotny PJ, et al. Longitudinal adverse event assessment in oncology clinical trials: the Toxicity over Time (ToxT) analysis of Alliance trials NCCTG N9741 and 979254. Lancet Oncol 2016; Advance online publication 12 April. DOI:

Mendoza T. Time for better presentation and analysis of adverse events. Lancet Oncol 2016; Advance online publication 12 April. DOI:

medwireNews ( ) is an independent medical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2016