Page 87 - Personalised medicine of fluoropyrimidines using DPYD pharmacogenetics Carin Lunenburg

toxicity: a
included in the meta-analysis of Terrazzino 2013
patients with cancer
systematic review and
(Morel 2006 and Deenen 2011).
5 of the 8 studies in this meta-analysis are also
treated with fluoropyrimidines.’
review and meta-analysis
5 of the 8 studies in this meta-analysis are also included separately in this risk analysis: Morel 2006,
fluoropyrimidines.’
meta-analysis of individual
included separately in this risk analysis: Morel 2006, Deenen 2011, Lee 2014, Rosmarin 2014 and
of individual patient data.
Deenen 2011, Lee 2014, Rosmarin 2014 and Meulendijks 2017.
patient data. Lancet Oncol
Meulendijks 2017.
If possible, a RR was calculated for each study based
Lancet Oncol 2015;16:1639
If possible, a RR was calculated for each study based on individual patient data and adjusted for age, sex,
2015;16:1639 -50. PubMed
on individual patient data and adjusted for age, sex, and treatment regimen. For 2 of the 5 studies for *13,
-50. PubMed PMID:
and treatment regimen. For 2 of the 5 studies for *13, it was not possible to use individual patient data. A
Supplement
toxicity: a systematic
included in the meta-analysis of Terrazzino 2013 (Morel 2006 and Deenen 2011).
patients with cancer treated with
PMID: 26603945.
it was not possible to use individual patient data. A random-effects model was used for the meta-
4
26603945.
random-effects model was used for the meta- analysis.
analysis.
Haematological toxicity included thrombocytopenia,
Haematological toxicity included thrombocytopenia, neutropenia, leukocytopenia, and anaemia. Gastro-
neutropenia, leukocytopenia, and anaemia. Gastro- intestinal toxicity included diarrhoea,
intestinal toxicity included diarrhoea, mucositis/stomatitis, and nausea/vomiting.
mucositis/stomatitis, and nausea/vomiting. Short timeframe was defined as shorter than the
Short timeframe was defined as shorter than the complete treatment duration, long timeframe as the
complete treatment duration, long timeframe as the whole treatment duration.
whole treatment duration.
In addition, a meta-analysis of 3 case-control studies
In addition, a meta-analysis of 3 case-control studies with in total 799 patients was performed for
with in total 799 patients was performed for c.1236G>A. One of these case-control studies is also
c.1236G>A. One of these case-control studies is also included in the meta-analysis of Rosmarin 2014
included in the meta-analysis of Rosmarin 2014 (Schwab 2008) and two in the meta-analysis of
(Schwab 2008) and two in the meta-analysis of Terrazzino 2013 (Schwab 2008 and Kleibl 2009). One
Terrazzino 2013 (Schwab 2008 and Kleibl 2009). One of these case-control studies is also included
of these case-control studies is also included separately in this risk analysis (Schwab 2008).
separately in this risk analysis (Schwab 2008).
Results:
Results:
Risk of grade ≥ 3 toxicity for *1/*13 compared to
RRadj (95% CI) RRadj (95% CI)
Risk of grade ≥ 3 toxicity for *1/*13 compared to
*1/*1: *1/*1:
incidence
any toxicity any toxicity haematological haematological toxicity
4.40 (2.08-9.30) (S) 4.40 (2.08-9.30) (S) 9.76 (3.03-31.48) 9.76 (3.03-31.48) (S)
patients) 22%
toxicity gastrointestinal
(S)
5.72 (1.40-23.33)
gastrointestinal toxicity
5.72 (1.40-23.33) (S)
toxicity hand-foot
(S)
- (RR could not be
hand-foot syndrome
- (RR could not be calculated due to
syndrome
calculated due to an incidence of 0%
an incidence of 0% in *1/*13)
in *1/*13)
The heterogeneity between the studies was
The heterogeneity between the studies was significant and substantial, possibly because of the
significant and substantial, possibly because of the small number of *1/*13.
small number of *1/*13.
There was no indication of publication bias.
There was no indication of publication bias. The results for any toxicity were similar when
The results for any toxicity were similar when patients carrying *2A and/or c.2846A>T were
patients carrying *2A and/or c.2846A>T were excluded from the meta-analysis. The association
excluded from the meta-analysis. The association remained significant with p< 0.0167 after
remained significant with p< 0.0167 after exclusion of any study from the meta-analysis,
exclusion of any study from the meta-analysis,
incidence for *1/*1
for *1/*1 (% of
(% of patients)
22%
table continues
85