Vous êtes sur la page 1sur 7

original articles Annals of Oncology

Annals of Oncology 23: 1986–1992, 2012

Published online 6 March 2012

The effect of guideline-consistent antiemetic therapy

on chemotherapy-induced nausea and vomiting
(CINV): the Pan European Emesis Registry (PEER)
M. Aapro1*, A. Molassiotis2, M. Dicato3, I. Peláez4, Á. Rodríguez-Lescure5, D. Pastorelli6, L. Ma7*,
T. Burke7, A. Gu7, P. Gascon8 & F. Roila9 on behalf of the PEER investigators
Medical Oncology and Radiation, IMO Clinique de Genolier, Genolier, Switzerland; 2School of Nursing Midwifery and Social Work, University of Manchester,
Manchester, UK; 3Hematology-Oncology, Luxembourg Medical Center, Luxembourg, Luxembourg; 4Hospital de Cabuenes, Gijón, Spain; 5Medical Oncology, Hospital
General Universitario de Elche, Elche, Spain; 6Oncologic Institute of the Veneto, Padova, Italy; 7Global Health Outcomes, Merck Sharp & Dohme Corp., Whitehouse
Station, USA; 8Institute of Hematology and Medical Oncology, IDIBAPS, University of Barcelona, Barcelona, Spain; 9Medical Oncology, Santa Maria Hospital, Terni, Italy

Received 2 November 2011; revised 6 January 2012; accepted 13 January 2012

Background: While guidelines for preventing chemotherapy-induced nausea and vomiting (CINV) are widely available,
clinical uptake of guidelines remains low. Our objective was to evaluate the effect of guideline-consistent CINV
prophylaxis (GCCP) on patient outcomes.
Patients and methods: This prospective, observational multicenter study enrolled chemotherapy-naive adults
initiating single-day highly or moderately emetogenic chemotherapy (HEC or MEC) for cancer. Patients completed 6-
day daily diaries beginning with cycle 1 for up to three chemotherapy cycles. The primary study end point, complete
response (no emesis and no use of rescue therapy) during 120 h after cycle 1 chemotherapy, was compared between
GCCP and guideline-inconsistent CINV prophylaxis (GICP) cohorts using multivariate logistic regression, adjusting for
potential confounding factors.
Results: In cycle 1 (N = 991), use of GCCP was 55 % and 46 % during acute and delayed phases, respectively, and
29 % for the overall study period (acute plus delayed phases). Complete response was recorded by 172/287 (59.9 % )
and 357/704 (50.7 % ) patients in GCCP and GICP cohorts, respectively (P = 0.008). The adjusted odds ratio for
complete response was 1.43 (95 % confidence interval 1.04–1.97; P = 0.027) for patients receiving GCCP versus GICP.
Conclusion: GCCP reduces the incidence of CINV after single-day HEC and MEC.
Key words: antiemetic therapy, chemotherapy, emesis, guidelines, nausea

introduction 30 % –90 % receiving moderately emetogenic chemotherapy

(MEC) experience CINV. The occurrence of CINV is typically
Several evidence-based consensus guidelines for preventing biphasic; thus, recommendations for antiemetic therapy are
chemotherapy-induced nausea and vomiting (CINV) are targeted to prevent CINV in the acute phase, occurring in the
published and regularly updated [1–3]. However, studies first 24 h, and the delayed phase, occurring > 24 h after
suggest that the clinical uptake of antiemetic guidelines is often chemotherapy [14–16]. Prevention of CINV from the start of
suboptimal, and CINV is a persistent problem for patients chemotherapy is important, both because delayed emesis is
receiving chemotherapy [4–8]. Patients who experience CINV correlated with the presence of acute emesis [17] and because
may be discouraged from completing their chemotherapy patients who experience CINV in one cycle are more likely to
regimen; CINV adversely impacts both quality of life and the develop anticipatory CINV during subsequent chemotherapy
ability to carry out the activities of daily living [9–11]. cycles [14].
Moreover, patients with emesis may require emergency care or There are a number of observational studies evaluating the
hospitalization, adding to the economic burden of cancer care effectiveness of different antiemetic regimens and the optimal
[12, 13]. means of implementing antiemetic guidelines in practice, but
In the absence of antiemetic therapy, > 90 % of patients the effect of guideline adherence in preventing CINV has not
receiving highly emetogenic chemotherapy (HEC) and been systematically studied [18–20]. The primary objective of
this prospective observational study was to analyze the
*Correspondence to: Dr M. Aapro, Clinique de Genolier, Institut Multidisciplinaire
proportions of patients with complete response (no emesis and
d’Oncologie IMO, Case Postale (PO Box) 100, Route du Muids 3, Genolier 1272, no use of rescue therapy) during the first 120 h after cycle 1
Switzerland. Tel: +41-223669106; Fax: +41-223669207; E-mail: maapro@genolier.net

© The Author 2012. Published by Oxford University Press on behalf of the European Society for Medical Oncology.
All rights reserved. For permissions, please email: journals.permissions@oup.com
Annals of Oncology original articles
HEC or MEC among patients receiving guideline-consistent electronic data capture form via double data entry. Data collected with
CINV prophylaxis (GCCP) as compared with guideline- regard to cancer chemotherapy included the intended and actual
inconsistent CINV prophylaxis (GICP). Our hypothesis was chemotherapy regimens, antiemetic therapy, and any rescue medications
that use of GCCP, as compared with GICP, would result in a used, administered, or prescribed (drug, dosage, and formulation for all).
greater proportion of patients achieving complete response in Patients completed the daily diary beginning before chemotherapy (on
the 120 h after the initiation of HEC or MEC. Secondary study day 1) and for a total of six calendar days. Before chemotherapy
administration, patients recorded whether they had any emesis in the prior
objectives were to describe the use of CINV prophylaxis for
24 h and, on three separate visual analogue scales (VAS, scored from 0,
HEC and MEC in European clinical practice and to describe
none, to 100, maximum), whether they had any nausea or anxiety in the
the experience of nausea and vomiting among patients
prior 24 h or any expectation of nausea. On subsequent days, items in the
receiving GCCP or GICP for HEC and MEC.
diary, with the exception of emetic events, were recorded with reference to
the prior 24 h, and study results were thus defined in 24-h increments
beginning at the time of initiation of chemotherapy: day 1 (acute phase)
patients and methods
was the 24-h period after the initiation of chemotherapy, and days 2–5
study design (delayed phase) were the 24- to 120-h period after initiation of
This prospective observational study was conducted from September 2009 chemotherapy. Patients recorded the time and date of emetic episodes, the
to June 2010 in eight European countries, including Austria, Belgium, severity of nausea and of anxiety (0–100 VAS), and use of rescue
France, Italy, Spain, Sweden, The Netherlands, and UK. medication (type and dosage). On day 6 of cycle 1, patients recorded use of
complementary therapies (such as acupressure, herbal remedies, yoga, and
relaxation techniques) and health care visits to manage nausea or vomiting
patients in the prior 5 days. Completion of the cycle 1 diary was required for
Male and female outpatients (aged ≥ 18 years) who were chemotherapy patients to continue in the study to cycle 2, and completion of the cycle 2
naive and scheduled to receive at least two cycles of single-day HEC or diary, to continue to cycle 3.
MEC were eligible for inclusion in the study. Intended treatment of cancer Chemotherapy regimens were categorized according to international
was with a minimum of any one of the following agents: cisplatin, guidelines based on the emetogenic potential of the agent in the regimen
cyclophosphamide, dacarbazine, oxaliplatin, carboplatin, doxorubicin, with the highest emetic risk [21]. Our study definition of guideline
epirubicin, irinotecan, oral temozolomide, oral vinorelbine. Patients were consistency of antiemetic therapy (GCCP) closely followed the 2006
excluded from the study if they received chronic systemic corticosteroid antiemetic guidelines of the Multinational Association of Supportive Care
therapy, concurrent abdominal or pelvic radiation therapy, or HEC or in Cancer (MASCC) [22] and was based, without imposing minimal doses,
MEC within 120 h (5 days) after day 1 chemotherapy administration. on whether recommended drugs were administered or prescribed on the
Other key exclusion criteria were the presence of brain metastases or recommended days (Table 1). Patients administered or prescribed the
vomiting in the 24 h before chemotherapy. recommended drugs on each day of the overall (acute and delayed phases)
period were classified as receiving GCCP for the overall phase. All other
patients were classified as receiving GICP during the overall phase.
The study was approved by the local ethics committee at each site, and
patients gave written informed consent. Adult patients initiating HEC or outcome assessments
MEC were recruited consecutively at study sites (hospital, community, and The primary study end point of complete response was defined as no
private oncology clinics). Patients completed 6-day daily diaries beginning emesis and no use of rescue therapy to relieve symptoms of nausea or
on the day of single-day chemotherapy and for one to three chemotherapy vomiting. No emesis was defined as no vomiting (expulsion of stomach
cycles. The study was designed specifically to avoid influencing treatment contents through mouth) and no retching (nonproductive attempts to
decisions or inducing any study-specific investigations. vomit). Distinct emesis episodes were separated by the absence of vomiting
Electronic data capture forms were used to record demographic and and retching for at least 1 min. No nausea was defined as nausea scored as
clinical data from source documents, including patient diaries, and to < 5 on the VAS. No CINV was defined as no emesis and no nausea. Rescue
complete investigator site questionnaires. Study sites anonymized any medication was defined as any medication taken in addition to
patient identifiers before forwarding information to be recorded on the prophylactic treatment for control of nausea or vomiting. Secondary

Table 1. Definition of guideline-consistent chemotherapy-induced nausea and vomiting prophylaxis (GCCP), based on MASCC 2006 guidelines [22]

Chemotherapy Acute phase (day 1) GCCP Delayed phase (days 2–4) GCCP
HEC Corticosteroid + NK1-RA + 5HT3-RA Corticosteroid days 2–4 + NK1-RA days 2–3
Female AC Corticosteroid + NK1-RA + 5HT3-RAa Corticosteroid +/or NK1-RA days 2–3c
MEC Corticosteroid + 5HT3-RAa,b Corticosteroid +/or 5HT3-RA days 2–3c
All 5HT3-RA were considered equivalent and interchangeable. Oral and i.v. formulations of corticosteroid, 5HT3-RA, and NK1-RA were considered
equivalent and interchangeable.
Palonosetron on day 1 was considered to provide prophylaxis on days 2–3 for MEC.
The study GCCP definition differs from MASCC 2006 guidelines in the delayed phase for female AC and MEC by permitting use of both listed agents
(+/or instead of or).
AC, anthracycline plus cyclophosphamide; HEC, highly emetogenic chemotherapy; MEC, moderately emetogenic chemotherapy; NK1-RA, neurokinin-1
receptor antagonist; 5HT3-RA, 5-hydroxytryptamine-3 receptor antagonist.

Volume 23 | No. 8 | August 2012 doi:10.1093/annonc/mds021 | 

original articles Annals of Oncology

antiemetic agents included any antiemetic other than the three primary Table 2. Patient demographic and clinical characteristics
classes of antiemetic agents (corticosteroids, neurokinin-1 receptor
antagonists [NK1-RAs], and 5-hydroxytryptamine-3 receptor antagonists Characteristic GCCP GICP Overall Pa
[5-HT3-RAs]). (N = 287) (N = 704) (N = 991)
Mean age, years (SD) 57.1 (11.7) 56.6 (11.2) 56.7 (11.4) 0.542
statistical analyses Female 225 (78.4) 497 (70.6) 722 (72.9) 0.012
Patient characteristics and study end points were summarized using Primary cancer diagnosis
descriptive statistics. Categorical and continuous variables were compared Breast 157 (54.7) 368 (52.3) 525 (53.0) < 0.001
between GCCP and GICP cohorts using Pearson’s χ2 test and Student’s t Colorectal 47 (16.4) 79 (11.2) 126 (12.7)
test, respectively. Multivariate logistic regressions were used to compare Lung 21 (7.3) 90 (12.8) 111 (11.2)
proportions of patients (GCCP versus GICP) achieving complete response Ovarian 20 (7.0) 31 (4.4) 51 (5.1)
during the study period of 120 h after initiation of cycle 1 chemotherapy Other 42 (14.6) 136 (19.3) 178 (18.0)
( primary end point), as well as other CINV outcomes, including no emesis, Metastatic disease 65 (22.6) 228 (32.4) 293 (29.6) 0.002
no nausea, and no CINV. The models included GCCP/GICP as a binary History of nausea or 79 (28.0) 141 (20.4) 220 (22.6) 0.010
variable and other relevant demographic and clinical variables. Multivariate vomiting
Poisson regression was used to compare the count of health care visits due Alcohol use (≥10 drinks per 22 (7.8) 83 (12.0) 105 (10.8) 0.056
to CINV in cycle 1 between the GCCP and GICP cohorts. Statistical week)
significance was assessed at the two-sided 0.05 level. Chemotherapy emetic risk category
The study was planned to have 80 % power to detect a 10-percentage HEC 21 (7.3) 168 (23.9) 189 (19.1) < 0.001
point between-group difference (α = 0.05, two-sided) in complete response Female AC 133 (46.3) 330 (46.9) 463 (46.7)
with 300 assessable patients in the GCCP cohort and 900 patients in the MEC 133 (46.3) 206 (29.3) 339 (34.2)
GICP cohort, assuming a 55 % complete response in the guideline- Pre-chemotherapy diary question responses (0–100 VAS), mean (SD)
consistent group and a 45 % complete response in the guideline- Pre-chemotherapy 27 (30) 32 (32) 30 (32) 0.037
inconsistent group. anxietyb
Expectation of nauseab 30 (31) 36 (32) 34 (32) 0.020

results Values represent number ( % ) of patients unless otherwise noted.

Chi-square test for categorical values; t test for continuous variables for
patients the comparisons between GCCP and GICP cohorts.
A total of 1128 eligible patients were enrolled from 52 study b
Measured during the 24-h period before start of chemotherapy.
sites; of these, 991 (87.9 % ), 888 (78.7 % ), and 769 (68.2 % ) AC, anthracycline plus cyclophosphamide; CINV, chemotherapy-induced
patients received single-day chemotherapy and completed nausea and vomiting; GCCP, guideline-consistent CINV prophylaxis;
diaries for chemotherapy cycles 1, 2, and 3, respectively. The GICP, guideline-inconsistent CINV prophylaxis; HEC, highly emetogenic
remaining 137 patients were excluded from study because they chemotherapy; MEC, moderately emetogenic chemotherapy; VAS, visual
did not complete the cycle 1 diary (n = 102), they did not analogue scale
receive intended chemotherapy (n = 26), or they received HEC
or MEC within 120 h after day 1 chemotherapy (n = 9).
Table 2 summarizes demographic and clinical characteristics
for patients included in the analyses for cycle 1. Patients
ranged in age from 19 to 84 years, with mean age of 57 years
in both cohorts. Almost three quarters of patients were women,
with a significantly greater proportion of women in the GCCP
cohort, and 99 % of patients were white. Breast cancer was the
most common diagnosis. Overall, one-fifth of patients received
HEC, almost half were women who received anthracycline plus
cyclophosphamide (female AC), and the remaining patients
(34 % ) received MEC (see Table 2).

antiemetic therapy
Antiemetic therapies prescribed for cycle 1 of single-day Figure 1. Prevalence of guideline-consistent CINV prophylaxis (GCCP)
chemotherapy are summarized in online Supplemental Table for cycle 1 single-day chemotherapy, by emetogenicity of chemotherapy
S1 (available at Annals of Oncology online) for the acute and and for the total study population. For the overall study period, patient
delayed phases according to guideline consistency. The use of assignment to the GCCP cohort was based on guideline consistency during
GCCP varied substantially between acute and delayed phases both acute and delayed phases (0–120 h post-chemotherapy). AC,
and among the three categories of chemotherapy emetogenicity anthracycline plus cyclophosphamide; HEC, highly emetogenic
(Figure 1). Of the emetic risk categories, HEC represented a chemotherapy; MEC, moderately emetogenic chemotherapy.
higher proportion of GICP than GCCP (23.9 % versus 7.3 % ),
while MEC was a higher proportion of GCCP than GICP MEC and higher in the acute than the delayed phase for
(46.3 % versus 29.3 % ; see Table 2). Guideline consistency was patients who received HEC or MEC. For female AC, guideline
highest overall (acute and delayed phases) for patients receiving consistency was higher in the delayed than the acute phase, in

 | Aapro et al. Volume 23 | No. 8 | August 2012

Annals of Oncology original articles
part, because we included as GCCP the prescription of Patients classified as GCCP had 1.43 times the odds of
corticosteroid and/or (rather than ‘or’) NK1-RA. complete response (95 % confidence interval 1.04–1.97; P =
The reasons for guideline inconsistency within the GICP 0.027) in the overall study period compared with patients
group varied across emetogenic risk group (see online receiving GICP, controlling for confounding factors (see
supplementalTable S2, available at Annals of Oncology online). Table 3). Similar statistically significant results were observed
For HEC GICP patients, non-prescribing of corticosteroid for the acute and delayed phases, though differences in CINV
(delayed phase, range from 16.1 % to 38.7 % ) and NK1-RA between the GCCP and GICP groups were larger for the acute
(acute phase, 51.8 % ) were the most common reasons for than delayed phase (online supplemental Table S4, available at
guideline inconsistency. For female AC and MEC, the primary Annals of Oncology online). The results for CINV end points
reasons for guideline inconsistent prescribing were non- during cycles 2 and 3 supported cycle 1 results (online
prescribing of NK1-RA use (acute phase, 7 % ) and supplemental Table S5, available at Annals of Oncology online).
corticosteroid and/or 5HT3-RA (delayed phase, 6.3 % to 29.6 The proportions of patients using health care resources
% ), respectively. during the 5 days after chemotherapy tended to be lower in the
Secondary antiemetic therapies were prescribed to 29 % of GCCP than the GICP cohort (Table 4). Visits to a general
patients overall with no significant difference between GCCP practitioner were the most frequent. Compared with those in
and GICP cohorts (31.0 % versus 28.1 % , P = 0.364; online the GICP cohort, patients in the GCCP cohort had
Supplemental Table S3, available at Annals of Oncology online). significantly fewer visits to specialists and the emergency room
Of the secondary antiemetic therapies, benzodiazepines were during the 5-day post-chemotherapy period.
prescribed significantly more frequently in the GCCP cohort
(10.5 % versus 1.0 % , P < 0.001) and benzamides, in the GICP
cohort (13.2 % versus 24.4 % , P < 0.001). The overall use of discussion
complementary therapies, most commonly herbal supplements
or special foods (e.g. ginger, fennel seeds) to prevent or Among chemotherapy-naive patients who received HEC or
manage vomiting or nausea, was similar in the two cohorts MEC in this large prospective study, the proportion with
(16.7 % versus 18.3 % , P = 0.55). complete response (no emesis or rescue therapy) during the
first 120 h after cycle 1 was significantly higher among those
receiving GCCP than those receiving GICP. Overall, almost 10
% more patients receiving GCCP achieved complete response
CINV events as compared with GICP. An improvement of 10 % in complete
The percentage of patients with complete response (no emesis response is considered to be a clinically meaningful difference
or rescue therapy) was significantly higher in the GCCP cohort [21]. The beneficial effect of GCCP in promoting complete
than in the GICP cohort during both acute and delayed phases response persisted after controlling for confounding factors.
as well as overall (during 120 h post-chemotherapy; Table 3). There was a consistent benefit of GCCP across CINV end
Moreover, the proportions of patients with other desirable points, and the results for cycles 2 and 3 were consistent with
CINV outcomes (no emesis, no nausea, and no CINV) were those for cycle 1. Guideline consistency in the acute phase of
higher in the GCCP than the GICP cohort in all phases, with cycle 1 was highly significant in reducing CINV, including
the differences either being statistically significant or suggesting nausea, in the acute phase: in multivariate analyses, adjusted
a strong trend. For all CINV end point results, the differences odds ratios for absence of CINV were significantly higher for
between GCCP and GICP cohorts were numerically greater in the GCCP cohort. The benefits of GCCP tended to be less
the acute than the delayed phase (see Table 3). pronounced in the delayed phase, although the odds ratio for

Table 3. Chemotherapy-induced nausea and vomiting (CINV) outcomes Table 4. Health care visits to manage CINV over 5 days after initiation of
by guideline consistency (GCCP versus GICP) in cycle 1 overall phase: chemotherapy—cycle 1
unadjusted and adjusted results
Parameter GCCP GICP Multivariate modelb
a b
GCCP GICP P Multivariate model (N = 287) (N = 704) Rate ratio (95 % CI) P
(N = 287) (N = 704) Odds ratio P GP visit 7 (2.4)a 29 (4.1) 0.68 (0.45–1.02) 0.062
Complete response 172 (59.9) 357 (50.7) 0.008 1.43 (1.04–1.97) 0.027 Specialist visit 3 (1.1) 11 (1.6) 0.51 (0.34–0.79) 0.002
No emesis 182 (63.4) 412 (58.5) 0.154 1.18 (0.86–1.63) 0.301 ER visit 4 (1.4) 12 (1.7) 0.57 (0.38–0.84) 0.004
No nausea 138 (48.1) 286 (40.6) 0.031 1.37 (0.99–1.90) 0.056 Hospital days 5 (1.7) 10 (1.4) 1.22 (0.79–1.86) 0.364
No CINV 122 (42.5) 242 (34.4) 0.016 1.41 (1.01–1.96) 0.041 a
Values represent total number of visits (normalized to number of visits
Chi-square test. per 100 patients).
b b
Model adjusted for age, sex, pre-chemotherapy nausea, pre-chemotherapy P values derived from Poisson regression models, accounting for
anxiety, expectation of nausea, use of primary antiemetic therapy not overdispersion, adjusting for the same set of confounding factors as those
recommended by guidelines, underdosing of primary antiemetic therapy, applied in the multivariate analyses depicted in Table 3.
and use of secondary antiemetic agents. CINV, chemotherapy-induced nausea and vomiting; ER, emergency room;
GCCP, guideline-consistent CINV prophylaxis; GICP, guideline- GCCP, guideline-consistent CINV prophylaxis; GICP, guideline-
inconsistent CINV prophylaxis. inconsistent CINV prophylaxis; GP, general practitioner.

Volume 23 | No. 8 | August 2012 doi:10.1093/annonc/mds021 | 

original articles Annals of Oncology

the primary end point, complete response, was significantly reflect economic constraints of hospitals and government
greater in the GCCP cohort. These findings parallel published payers considering the higher costs of branded-only antiemetic
observations that, in routine practice, nausea and delayed therapies. Nonetheless, this is an area especially needing
symptoms remain the most challenging to manage [11, 23–26]. improvement, and providing appropriate CINV prophylaxis to
The use of GCCP was relatively low overall (29 % ) and the subgroups of patients at highest risk of CINV will likely
lowest (11 % ) for patients in the highest emetic risk category, have the greatest impact on improving patient outcomes.
most commonly because of the less than optimal use of NK1- Comprehensive and long-term efforts consisting of efficient
RA and corticosteroids, particularly out to day 4. education, training, and monitoring of all individuals involved
Patients receiving GCCP had significantly fewer specialist are needed to achieve better adherence to antiemetic guidelines
and emergency room visits to manage CINV over the 5 days [6, 7]. Key components of successful prior multifaceted
after chemotherapy than those receiving GICP and marginally strategies to improve guideline adherence include educational
nonsignificantly fewer general practice visits (P = 0.06). While outreach visits by opinion leaders [18, 20], the use of
the absolute numbers of hospital days and clinic and standardized antiemetic protocols included in chemotherapy
emergency visits due to CINV were not large, the differences order forms [19], providing feedback to clinicians on the
between cohorts could have important economic and clinical extent and severity of patient CINV outcomes [5], and clinical
implications considering the many patients treated with interventions by pharmacists in the event of inappropriate
multiple cycles of HEC and MEC regimens on an annual basis antiemetic orders [20].
throughout Europe. Our study results indicate that there is a significant benefit of
This large prospective study has enabled us to examine guideline-consistent antiemetic therapy across a range of CINV
antiemetic prescribing patterns across a range of European end points in the acute and delayed phases and overall. The
countries and oncology centers. The sample size, consisting of incidence of CINV was relatively high, and the prevalence of
nearly 1000 assessable patients treated with MEC or HEC, GCCP relatively low, for the overall study period. Trends favored
provides a robust evaluation of the benefits of guideline- the GCCP group for lower health care resource use secondary to
consistent prophylaxis of CINV, including the newest CINV. The results of this observational study highlight the need
antiemetic therapies, aprepitant and palonosetron. As a result, to improve the transferability to clinical practice of antiemetic
the study results will generalize well to current Western research. Studies are needed to explore the barriers to guideline
European clinical practice. use for clinicians and to test the effects of different strategies to
A weakness of the study is that the GCCP classification was enhance their use of guidelines. Strategies targeting patients
based on CINV prophylaxis as prescribed for oral agents, receiving HEC or females receiving AC regimens, where
rather than on actual use by patients. Therefore, patients who adherence is lower and outcomes poorer are recommended.
failed to adhere to the prescribed regimen would have been Education on guideline recommendations for corticosteroids
misclassified, likely leading to an underestimation of the (primarily delayed phase) and NK1-RA (overall phase) within
differences between GCCP and GICP. Moreover, we cannot these subgroups is also recommended. Better communication
rule out the possibility of omissions or errors by patients in the and implementation of antiemetic guidelines must be
diaries. Other study weaknesses include those common to considered as a means to reduce the burden of CINV.
observational studies, namely, cohort assignment based on
physician prescribing rather than randomization, resulting in
baseline differences between study cohorts and potential
confounding by indication. The use of multivariate logistic We thank all the investigators who participated in this study
regression addresses the latter limitation; however, statistically ( please see Appendix for listing of Chief Investigators by
significant differences could occur due to chance. study site).
Comparisons with previous studies on the prevalence and
effect of guideline consistency are complicated by differing
guidelines, definitions of consistency, time periods, locations,
and patient populations. Adherence to guidelines in earlier This work was supported by Merck Sharp & Dohme Corp.
single-center studies in the United States and Europe was Medical writing and editorial assistance was provided by
reportedly better for the acute than the delayed phase [4, 5, 8], Elizabeth V. Hillyer and funded by Merck Sharp & Dohme
consistent with our findings for the total study population. Corp., a subsidiary of Merck & Co., Inc., Whitehouse
Similarly, earlier studies that evaluated the effect of guideline Station, NJ.
implementation reported better antiemetic outcomes as
guideline adherence increased; these were smaller single-center
studies based in the United States before the availability of
aprepitant and palonosetron [5, 19, 20]. MA holds no shares in any drug company or in mutual funds
The emetogenicity of chemotherapy is accepted as the most investing exclusively in pharma; he has received study grants,
important risk factor for CINV and is used to guide selection honoraria for consultancy, and/or is on the speaker bureau of
of antiemetic therapy [14]. Our findings provide support for Abraxis, Amgen, AstraZeneca, Bayer Schering, Bristol-Myers,
the use of GCCP to reduce the incidence of CINV after cycle 1 Celgene, Cephalon, GSK, Helsinn, Hospira, Johnson and
HEC, female AC, and MEC. The low prevalence of GCCP in Johnson Ortho Biotech, Merck, MSD, Novartis, Pfizer,
the highest emetogenic risk regimens (11 % for HEC) may Pierre-Fabre, Roche, Sandoz, Schering, sanofi-aventis, Vifor.

 | Aapro et al. Volume 23 | No. 8 | August 2012

Annals of Oncology original articles
AM has served in the advisory board and received honoraria 19. Nolte MJ, Berkery R, Pizzo B et al. Assuring the optimal use of serotonin
and research funding from Merck & Co Ltd; received antagonist antiemetics: the process for development and implementation of
institutional antiemetic guidelines at Memorial Sloan-Kettering Cancer Center. J
honoraria and research funding from Roche UK Ltd; research
Clin Oncol 1998; 16: 771–778.
funding from Acacia Pharma; and honoraria from ProStrakan
20. Dranitsaris G, Leung P, Warr D. Implementing evidence based antiemetic
and Bayer. MMD is a member of the Merck speakers bureau guidelines in the oncology setting: results of a 4-month prospective intervention
and advisory board. IP, ÁR-L, DP, and AG have no conflicts to study. Support Care Cancer 2001; 9: 611–618.
declare. PG is a member of a Merck advisory board. LM was 21. Multinational Association of Supportive Care in Cancer (MASCC). MASCC/ESMO
an employee of Merck at the time of the study. TB is an Antiemetic Guidelines 2011. http://www.mascc.org/mc/page.do?sitePageId =
employee of Merck. FR has served on the advisory board and 88041 [12 October 2011, date last accessed].
received honoraria from Merck. All remaining authors have 22. Roila F, Hesketh PJ, Herrstedt JAntiemetic Subcommittee of the Multinational
declared no conflicts of interest. Association of Supportive Care in CancerPrevention of chemotherapy- and
radiotherapy-induced emesis: results of the 2004 Perugia International
Antiemetic Consensus Conference. Ann Oncol 2006; 17: 20–28.
references 23. Molassiotis A, Saunders MP, Valle J et al. A prospective observational study of
chemotherapy-related nausea and vomiting in routine practice in a UK cancer
1. Roila F, Herrstedt J, Aapro M et al. Guideline update for MASCC and ESMO in the centre. Support Care Cancer 2008; 16: 201–208.
prevention of chemotherapy- and radiotherapy-induced nausea and vomiting: results 24. Booth CM, Clemons M, Dranitsaris G et al. Chemotherapy-induced nausea and
of the Perugia consensus conference. Ann Oncol 2010; 21Suppl 5: v232–v243. vomiting in breast cancer patients: a prospective observational study. J Support
2. National Comprehensive Cancer Network (NCCN). Antiemesis. NCCN Guidelines Oncol 2007; 5: 374–380.
2011; Version 3 2011; http://www.nccn.org/professionals/physician_gls/PDF/ 25. Grunberg SM, Deuson RR, Mavros P et al. Incidence of chemotherapy-induced
antiemesis.pdf (12 October 2011, date last accessed). nausea and emesis after modern antiemetics. Cancer 2004; 100: 2261–2268.
3. Basch E, Prestrud AA, Hesketh PJ et al. Antiemetics: american Society of Clinical 26. Roscoe JA, Morrow GR, Hickok JT, Stern RM. Nausea and vomiting remain a
Oncology clinical practice guideline update. J Clin Oncol 2011; 29: 4189–4198. significant clinical problem: trends over time in controlling chemotherapy-induced
4. Fabi A, Barduagni M, Lauro S et al. Is delayed chemotherapy-induced emesis well nausea and vomiting in 1413 patients treated in community clinical practices. J
managed in oncological clinical practice? An observational study. Support Care Pain Symptom Manage 2000; 20: 113–121.
Cancer 2003; 11: 156–161.
5. Mertens WC, Higby DJ, Brown D et al. Improving the care of patients with regard to
chemotherapy-induced nausea and emesis: the effect of feedback to clinicians on
adherence to antiemetic prescribing guidelines. J Clin Oncol 2003; 21: 1373–1378. appendix
6. Grunberg SM. Obstacles to the implementation of antiemetic guidelines. J Natl
Compr Canc Netw 2009; 7: 601–605. Chief Investigators for the Pan European Emesis Registry
7. Kaiser R. Antiemetic guidelines: are they being used?. Lancet Oncol 2005; 6: (PEER) are listed below.
622–625. Austria: Paul Sevelda (Vienna General Hospital), Herbert
8. Burmeister H, Aebi S, Studer C et al. Adherence to ESMO clinical Watzke (Medical University Vienna), Guntram Winder
recommendations for prophylaxis of chemotherapy-induced nausea and vomiting.
(Hospital Dornbirn), Richard Greil (University Hospital
Support Care Cancer 2011 January 14, [epub ahead of print], doi:10.1007/
Salzburg), Alois Lang (Hospital Rankweil); Belgium: Simon
9. Bloechl-Daum B, Deuson RR, Mavros P et al. Delayed nausea and vomiting Van Belle (UZ Gent), Dirk Schrijvers (ZNA Middelheim),
continue to reduce patients’ quality of life after highly and moderately emetogenic Didier Verhoeven (AZ KLINA), Wim Demey (AZ Sint-Jozef ),
chemotherapy despite antiemetic treatment. J Clin Oncol 2006; 24: 4472–4478. Francoise Majois (Centre Hospitalier Jolimont La Louvière),
10. Cohen L, de Moor CA, Eisenberg P et al. Chemotherapy-induced nausea and Frederique Bustin (CHR de la Citadelle Liège), Chantal Andre
vomiting: incidence and impact on patient quality of life at community oncology (CHR de la Citadelle Liège); France: Marc Espie (Hôpital Saint
settings. Support Care Cancer 2007; 15: 497–503.
-Louis), Maya Hacini (Hôpital Chambery), Anne-Claire Hardy
11. Glaus A, Knipping C, Morant R et al. Chemotherapy-induced nausea and vomiting in
Bessard (Clinique Armoricaine), Nicolas Isambert (Centre G-F
routine practice: a European perspective. Support Care Cancer 2004; 12: 708–715.
12. Burke TA, Wisniewski T, Ernst FR. Resource utilization and costs associated with Leclerc), Alain Lortholary (Centre Catherine de Sienne),
chemotherapy-induced nausea and vomiting (CINV) following highly or moderately Pierre-Jean Souquet (CH Lyon Sud), Mohamed Hebbar
emetogenic chemotherapy administered in the US outpatient hospital setting. (CHRU Lille-hôpital Claude Huriez); Italy: Massimiliano
Support Care Cancer 2011; 19: 131–140. Berretta (Centro di Riferimento Oncologico IRCCS), Claudio
13. Ihbe-Heffinger A, Ehlken B, Bernard R et al. The impact of delayed Dazzi (Presidio Ospedaliero S. Maria delle Croci), Manlio
chemotherapy-induced nausea and vomiting on patients, health resource Ferrarini (Istituto Nazionale per la Ricerca sul Cancro IRCCS),
utilization and costs in German cancer centers. Ann Oncol 2004; 15: 526–536.
Marzia Locatelli (Istituto Europeo di Oncologia IRCCS), Vito
14. Hesketh PJ. Chemotherapy-induced nausea and vomiting. N Engl J Med 2008;
358: 2482–2494. Lorusso (Presidio Ospedaliero Vito Fazzi, Polo Oncologico
15. Feyer P, Jordan K. Update and new trends in antiemetic therapy: the continuing Giovanni Paolo II), Paolo Marchetti (Azienda Ospedaliera
need for novel therapies. Ann Oncol 2011; 22: 30–38. S. Andrea), Andrea Martoni (Azienda Ospedaliero-
16. Herrstedt J. Antiemetics: an update and the MASCC guidelines applied in clinical Universitaria Policlinico S.Orsola-Malpighi), Davide Pastorelli
practice. Nat Clin Pract Oncol 2008; 5: 32–43. (Istituto Oncologico Veneto IRCCS), Sandro Pignata
17. Warr DG, Grunberg SM, Gralla RJ et al. The oral NK(1) antagonist aprepitant for (Fondazione Giovanni Pascale, Istituto Nazionale per lo Studio
the prevention of acute and delayed chemotherapy-induced nausea and vomiting: e la Cura dei Tumori IRCCS), Monica Giordano (Az.
pooled data from 2 randomised, double-blind, placebo controlled trials. Eur J
Cancer 2005; 41: 1278–1285.
Ospedaliera Sant’Anna); Netherlands: Joachim Aerts (Amphia
18. Roila F. Transferring scientific evidence to oncological practice: a trial on the Zkhs, Loc Molengracht), Reinoud Blaisse (Alysis Zorggroep,
impact of three different implementation strategies on antiemetic prescriptions. Loc Rijnstate), Agnis van der Wouw (VieCuri MC Noord-
Support Care Cancer 2004; 12: 446–453. Limburg); Spain: Ignacio Peláez (Hospital de Cabueñes), Silvia

Volume 23 | No. 8 | August 2012 doi:10.1093/annonc/mds021 | 

original articles Annals of Oncology

Antolin Novoa (Hospital Teresa Herrera—Complejo Manchester & Christie Hospital NHS Foundation Trust),
Hospitalario Universitario A Coruña-CHUAC), Jose Esteban Elaine Rankin (Ninewells Hospital and Medical School), Leslie
Salgado Pascual (Hospital de Navarra), Vicente Alberola Samuel (Aberdeen Royal Infirmary), Anirban Chatterjee (Royal
Candel (Hospital Arnau de Vilanova), Álvaro Rodríguez- Shrewsbury Hospital), Sarah Khan (Nottingham University
Lescure (Hospital General Universitario de Elche), Laura Hospital and King’s Mill Hospital), Neville Davidson
Palomar (Hospital Universitario La Fe), Virtudes Soriano (Broomfield Hospital); Sweden: Max Flogegard
Teruel (Instituto Valenciano de Oncologia); UK: Mike Bayne (Blodmottagningen); Bengt Norberg (Onkolog Kliniken), Petra
(Dorset County Hospital), Karen McAdam (Peterborough and Flygare (Onkolog Kliniken #9097), Bo Nordenskjold
Stamford Hospitals), Alex Molassiotis (The University of (Onkologkliniken).

Annals of Oncology 23: 1992–1998, 2012

Published online 13 June 2012

Daily skin care habits and the risk of skin eruptions

and symptoms in cancer patients
H. J. Byun1,2,3, H. J. Lee4,5*, J. I. Yang6, K. H. Kim1,2,3, K. O. Park7, S. M. Park7,8, K. E. Lee5,9,
J. Choi10, D.-Y. Noh5,11 & K. H. Cho1,2,3*
Skin Cancer/Chemotherapy Skin Care Center, Seoul National University Cancer Hospital, Seoul;; 2Department of Dermatology, Seoul, National University College of
Medicine, Seoul;; 3Institute of Dermatological Science, Medical Research Center, Seoul National University, Seoul;; 4Gastric Cancer Center, Seoul National University
Cancer Hospital, Seoul;; 5Department of Surgery and Cancer Research Institute, Seoul National University College of Medicine, Seoul;; 6Department of Internal Medicine,
Healthcare Research Institute, Seoul National University Hospital Gangnam Healthcare Center, Seoul;; 7Center for Cancer Education and Information, Seoul National
University Cancer Hospital and Department of Nursing, Seoul National University Hospital, Seoul;; 8Department of Family Medicine, Seoul National University College of
Medicine, Seoul;; 9Thyroid Cancer Center, Seoul National University Cancer Hospital, Seoul;; 10Department of Biomedical Engineering, Seoul National University College
of Medicine, Seoul;; 11Breast Cancer Center Seoul National University Cancer Hospital, Seoul, South Korea

Received 15 November 2011; revised 29 March 2012; accepted 10 April 2012

Background: Cancer patients are at high risk for skin problems because rapidly proliferating skin cells are susceptible
to anticancer therapies. However, the effects of daily skin care habits on development of skin problems in cancer
patients have rarely been studied.
Patients and methods: We conducted a survey of daily skin care habits and the presence of skin problems in 866
cancer patients.
Results: Hot water bath >1 h significantly increased the risk of definite eruptions [odds ratio (OR) 4.09] and the risk of
itching or pain on the skin (OR 1.73). Diligent use of moisturizers did not decrease the risk of definite eruptions and
symptoms, and daily bathing, scrubbing off the skin while bathing, and sun protection did not influence the risk of
definite eruptions and symptoms. Subgroup analysis of 183 breast cancer patients showed results similar to the total
results, including that hot water bath >1 h significantly increased the risk of definite eruptions (OR 3.41).
Conclusions: Being a cross-sectional study, our study could not prove causality. However, at the present stage of
knowledge, avoidance of hot water baths of protracted duration should be first emphasized in patient education to
prevent skin problems in cancer patients.
Key words: bath, cancer, eruptions, itching, moisturizer, skin care habits

introduction cytotoxic agents and radiation therapy. To overcome the

Cancer patients are at high risk for skin problems because nonspecific damage of normal cells by cytotoxic agents,
rapidly proliferating skin cells are susceptible to anticancer targeted agents have been developed and are increasingly used;
however, the most widely used targeted agents are epidermal
*Correspondence to: Prof. H. J. Lee, Department of Surgery and Cancer Research
growth factor receptor (EGFR) inhibitors, which notoriously
Institute, Seoul National University College of Medicine, 101 Daehang-ro, Jongno-gu, induce acneiform rash in ∼45%–100% [1]. Skin lesions can
Seoul 110-744, South Korea. Tel: +82-2-766-3975; Fax: +82-2-766-3975; cause cosmetic problems leading to impairment in the quality
E-mail: appe98@snu.ac.kr; Prof. K. H. Cho, Department of Dermatology, Seoul National
of life and poor patient compliance, more dose delays, and
University College of Medicine, 101 Daehang-ro, Jongno-gu, Seoul 110-744, South
Korea. Tel: +82-2-2072-2412; Fax: +82-2-742-7344; E-mail: khcho@snu.ac.kr even discontinuation of antineoplastic therapy [2, 3].

© The Author 2012. Published by Oxford University Press on behalf of the European Society for Medical Oncology.
All rights reserved. For permissions, please email: journals.permissions@oup.com.