original articles
21. Cooper R, Mason M, Finan P et al. Defunctioning stomas prior to
chemoradiation for anal cancer are usually permanent. Colorectal Dis 2012;
14(1): 87–91.
22. Northover J, Glynne-Jones R, Sebag-Montefiore D et al. Chemoradiation for the
treatment of epidermoid anal cancer: 13-year follow-up of the first randomised
UKCCCR Anal Cancer Trial (ACT I). Br J Cancer 2010; 102(7): 1123–8.
Annals of Oncology
23. Aggarwal A, Gayadeen S, Robinson D et al. Clinical target volumes in anal cancer:
calculating what dose was likely to have been delivered in the UK ACT II trial
protocol. Radiother Oncol 2012; 103(3): 341–6.
24. Buettner F, Gulliford SL, Webb S et al. The dose–response of the anal sphincter
region—an analysis of data from the MRC RT01 trial. Radiother Oncol 2012; 103
(3): 347–52.
Annals of Oncology 25: 1622–1628, 2014
doi:10.1093/annonc/mdu189
Published online 14 May 2014
ABVD (8 cycles) versus BEACOPP (4 escalated cycles ≥4
baseline): final results in stage III–IV low-risk Hodgkin
lymphoma (IPS 0–2) of the LYSA H34 randomized trial†
N. Mounier1*, P. Brice2, S. Bologna3, J. Briere2, I. Gaillard4, M. Heczko5, J. Gabarre6,
O. Casasnovas7, J. Jaubert8, P. Colin9, A. Delmer10, A. Devidas11, E. Bachy12, E. Nicolas-Virelizier13,
A. Aoudjhane14, C. Humbrecht15, M. Andre16 & P. Carde17, on behalf of the Lymphoma Study
Association (LYSA)
1
Department of Onco-Hematology, Archet Hospital, Nice; 2Department of Hematology, Saint-Louis Hospital, Paris; 3Department of Hematology, Nancy Hospital;
Department of Hematology, Henri Mondor Hospital, Creteil; 5Department of Hematology, Besancon Hospital; 6Department of Hematology, la Pitié-Salpétrière Hospital,
Paris; 7Department of Hematology, CHU de Dijon Hospital, Dijon; 8Department of Hematology, St Etienne Hospital, St Etienne; 9Department of Hematology, Courlancy
Cancer Institute, Reims; 10Department of Hematology, Reims Hospital, Reims; 11Department of Hematology, Corbeil Hospital, Corbeil Essonnes; 12Department of
Hematology, Lyon Sud Hospital, Pierre Benite; 13Department of Hematology, Leon Berard Cancer Center, Lyon; 14Department of Hematology, St Antoine Hospital, Paris;
15
Department of Hematology, Colmar Hospital, Colmar, France; 16Department of Hematology, Mont Godine Hospital, Yvoire, Belgium; 17Department of Hematology,
Gustave Roussy Institute, Villejuif, France
4
Received 5 February 2014; revised 19 March 2014; accepted 5 May 2014
Background: Treatment with escalated BEACOPP achieved a superior time to treatment failure over ABVD in patients
with disseminated Hodgkin lymphoma. However, recent clinical trials have failed to confirm BEACOPP overall survival
(OS) superiority over ABVD. In addition, the gain in low-risk patients is still a matter of debate.
Patients and methods: We randomly compared ABVD (8 cycles) with BEACOPP (escalated 4 cycles ≥baseline 4 cycles)
in low-risk patients with an International Prognostic Score (IPS) of 0–2. The primary end point was event-free survival (EFS).
This parallel group, open-label phase 3 trial was registered under #RECF0219 at French National Cancer Institute.
Results: One hundred and fifty patients were randomized in this trial (ABVD 80, BEACOPP 70): 28 years was the median
age, 50% were male and IPS was 0–1 for 64%. Complete remission rate was 85% for ABVD and 90% for BEACOPP.
Progression or relapses were more frequent in the ABVD patients than in the BEACOPP patients (17 versus 5 patients). With a
median follow-up period of 5.5 years, seven patients died: six in the ABVD arm and one in the BEACOPP arm (HL 3 and 0,
2nd cancer 2 and 1, accident 1 and 0). The EFS at 5 years was estimated at 62% for ABVD versus 77%, for BEACOPP
[hazards ratio (HR) = 0.6, P = 0.07]. The progression-free survival (PFS) at 5 years was 75% versus 93% (HR = 0.3, P = 0.007).
The OS at 5 years was 92% versus 99% (HR = 0.18, P = 0.06).
Conclusion: Fewer progressions/relapses were observed with BEACOPP, demonstrating the high efficacy of the more intensive regimen, even in low-risk patients. However, additional considerations, balancing treatment-related toxicity and late morbidity due to salvage may help with decision-making with regard to treatment with ABVD or BEACOPP.
Key words: Hodgkin lymphoma, intensive chemotherapy, phase 3 trial, prognostic factors, secondary malignancy
*Correspondence to: Prof. Nicolas Mounier, Onco-Hématology Department, Archet
Hospital, 151 route de Saint Antoine Ginestière, 06200 Nice, France. Tel: +33-4-92-0358-41; E-mail: [email protected]
†
This study was presented, in part, at the 12th International Conference on Malignant
Lymphoma, 19–22 June 2013, in Lugano, Switzerland, and at the 9th International
Symposium on Hodgkin Lymphoma, 12–15 October 2013, in Cologne, Germany.
© The Author 2014. Published by Oxford University Press on behalf of the European Society for Medical Oncology.
All rights reserved. For permissions, please email: [email protected].
original articles
Annals of Oncology
introduction
A majority of patients with Hodgkin lymphoma (HL) are now
being cured, even in the advanced stages. However, balancing the
risks and benefits, treatment strategies are still a matter of debate.
The standard approach to treatment uses doxorubicin, bleomycin,
vinblastine and dacarbazine (ABVD) as first-line therapy [1]. For
the 25% of patients who relapse after initial treatment, second-line
treatment with high-dose chemotherapy is followed by autologous
stem cell transplantation. The alternative approach aims to cure as
many patients as possible with aggressive first-line chemotherapy
using increased doses of bleomycin, etoposide, doxorubicin,
cyclophosphamide, vincristine, procarbazine and prednisone
(escalated BEACOPP) [2–4]. Accumulated data have shown better
progression-free survival (PFS) and overall survival (OS) rates, but
this alternative approach exposes patients to substantial acute
chemotherapy-related toxicity [5, 6].
Clinical prognostic indices may help to identify patients at a
low risk of treatment failure to be considered for standard treatment or to identify patients at a higher risk to be considered for
more intensive therapy or a combination with new drugs. For
more than 15 years, the International Prognostic Score (IPS),
which incorporates seven clinical parameters, has been the most
widely accepted risk stratification model [7]. However, for
patients treated in the ABVD era, IPS predicts for a much narrower range of outcomes than has been previously noted [8].
More specifically, the difference in PFS in patients with low-risk
and high-risk scores has diminished to 15%.
The intergroup trial 20012, led by EORTC, grouping LYSA
(formerly GELA), ALLG, NCRI LYG, GELCAB, NCIC and NLG
was a phase 3 trial to assess whether treatment with four cycles of
escalated BEACOPP (bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine and prednisone) followed
by four cycles of BEACOPP-baseline improved event-free survival
(EFS) when compared with the eight cycles of the standard ABVD
(doxorubicin, bleomycin, vinblastine, dacarbazine) regimen in
patients with measurable stage III or IV HL, with a high-risk IPS
(≥3)[9]. We present herein the results of a trial conducted in parallel to the intergroup 20012 trial in LYSA centers only, which
aimed to assess the EFS gain after stratification in low-risk patients
with an IPS ranging from 0 to 2 (registered under #RECF0219 at
French National Cancer Institute).
methods
patients
Patients were included in the study if they had histologically documented
HL (except lymphocyte predominant nodular type), were treatment naïve,
were 16–60 years old, were a clinical stage III or IV, had at least one bi-dimensionally measurable target lesion (even if extranodal only), had a WHO
performance status of 0–2, had an IPS of 0–2 and had an absence of other
malignancies (except basal cell skin and in situ uterine cervix carcinomas).
Patients were excluded from the study if they had an absence of measurable
disease, were of childbearing-potential without the use of effective contraception, were pregnant or lactating, had an active infection, had severe cardiopulmonary, neurological or metabolic disease that interfered with normal
life, or had inadequate liver (bilirubin ≤2.5×normal) or renal (creatinine
≤150 μmol/l or ≤2.0 mg/dl) function (unless the inadequate function was
due to HL).
Volume 25 | No. 8 | August 2014
study design and treatment
This was a parallel group, open-label randomized trial. We used computerassisted permuted-block randomization (block size of four, allocation ratio
1:1) to assign treatment. Randomization was stratified by participating
center. A statistician located centrally at the GELA research clinic supervised
the randomization procedure.
The doses and schedules of the chemotherapy combinations are listed in
supplementary Table S1 available at Annals of Oncology online. ABVD was
administered for eight cycles, depending on the response assessed at the end
of cycles 4 (C4), 6 (C6) and 8 (C8), using clinical and computed tomography
(CT) criteria. Positron emission tomography (PET) scans were not allowed.
BEACOPP was administered for eight cycles (escalated for four cycles, then
followed by baseline administration for four cycles). The checkpoints for
pursuing the treatment as planned were: achievement at the end of the
fourth cycle (i.e. the 22nd or 28th day, depending on the arm) of at least a
partial response (regression of more than 50%, i.e. PR >50%) and normalization of bone marrow biopsy in the case of initial involvement; achievement
at the end of C6 of at least a PR >75%. Patients in CR after four or six cycles
have to complete the scheduled eight ABVD or BEACOPP cycles. Any
protocol therapy discontinuation while the patient was in PR, progression,
or unknown status or was administered a new treatment was defined and
recorded as an event (treatment failure). No radiotherapy could be given as
part of the protocol treatment unless it was considered as an event. The
prophylactic administration of G-CSF and trimethoprim/sulfamethoxazole
was mandatory in the BEACOPP arm. The evaluation of response was
carried out according to the Response Evaluation Criteria for HL [10].
Local or national ethics committees approved the study protocol, according
to the laws of every country. The study was done in accordance with
the Declaration of Helsinki. Patients provided written informed consent
before inclusion. Hematologic and non-hematologic toxicities were graded
according to the National Cancer Institute Common Toxicity Criteria
(Version 2.0).
statistical analysis
The EFS was the primary end point, which was defined as the time from randomization to the early discontinuation of the protocol treatment, the
absence of complete remission (CR/CRu) after eight cycles or the occurrence
of relapse, progression or death. The study was designed for 150 patients to
be randomized over 5 years. It was determined that 26 events were required
to detect a 15% increase in the 5-year EFS rate from 75% (ABVD) to 90%
(BEACOPP), based on a one-sided log-rank test at error rates of α = 0.05 and
β = 0.20. Secondary end points were defined as CR/CRu, PFS, OS and the occurrence of second malignancies. We calculated hazard ratios (HR) by the
Cox proportional-hazards analysis. Dose intensity for chemotherapy drugs
was calculated according to Hryniuk and Goodyear [11]. We compared
patients’ qualitative variables with the χ 2 or Fisher’s exact test. We judged
differences significant if P-values were <0·05 (two-sided). All analyses were
conducted as intention-to-treat using SAS 9.2 software.
results
As stated in the Methods section, 150 eligible patients were enrolled in the study between February 2003 and August 2008.
Pathological characteristics were centrally reviewed in 148
patients (98%). Most of the patients (89%) had a nodular sclerosis subtype of HL. The median age of the patients was 28 years
(range: 16–60). The sex ratio was 1:1. Eighty-seven patients
(64%) presented with an IPS of 0–1. Seventy-two patients (48%)
presented with stage IV disease and 30 (22%) presented with a
tumor mass larger than 10 cm. Sixty-one patients (40%) had at
doi:10.1093/annonc/mdu189 | 
original articles
Annals of Oncology
least one extranodal site. The marrow was involved in 12
patients (8%). Eighty patients were randomized to the ABVD
arm and 70 were randomized to the BEACOPP arm. The demographic and baseline disease characteristics, according to treatment arm, are listed in Table 1.
Table 1. Baseline characteristics of the patients
Characteristics
(no. of patients)
ABVD
(n = 80)
Age (years)
Median
Range
<45
Sex
Male
Female
Erythrosite sedimentation rate
Above 30 mm
Below 30 mm
Performance status
0
1
2
Presence of general
symptoms
Ann Arbor stage
Stage III
Stage IV
Bone marrow involvement
International prognostic index
0
1
2
3
Missing
28
16–60
74
%
BEACOPP
(n = 70)
%
92
28
16–58
64
91
42
38
52
48
33
37
47
53
51
29
64
36
40
30
62
38
45
30
5
49
56
37
6
61
45
24
1
38
64
34
1
54
48
32
6
60
40
8
30
40
6
43
57
9
14
33
22
2
9
17
41
27
2
8
32
21
4
5
11
46
30
6
response to treatment
As shown in Figure 1, treatment started in 77 patients and 68
because three withdrew their consent and two presented staging
errors. Overall, 68 out of 77 completed the ABVD treatment
and 61 out of 68 completed BEACOPP. No deaths occurred due
to toxicity.
The main reasons for treatment discontinuation were treatment failure (3 and 2), patient refusal (3 and 2) and toxicity (1
and 3). In the ABVD arm, the toxicity occurred as one case of
intracardiac thrombosis. In the BEACOPP arm, there were two
infectious complications (colitis, pneumopathy) and one allergy
to etoposide.
At the end of the treatments, failure to achieve CR occurred
in 15 patients: 10 in the ABVD arm and 5 in the BEACOPP
arm. As shown in Table 2, the CR rate was 85% for ABVD and
90% for BEACOPP. After the end of treatment evaluation, three
non-CR patients were given consolidation radiotherapy in
ABVD ( protocol violation) and none in BEACOPP.
survival outcomes
The median follow-up period was 5.5 years. Seventeen relapses
were observed: 14 patients in the ABVD arm versus 3 in the
BEACOPP arm. The EFS at 5 years was estimated at 62% (50; 71)
versus 77% (65; 85), respectively [HR = 0.6 (0.33; 1.06), P = 0.07]
(Figure 2). Regarding the sample size computation hypothesis on
the primary end point, the one-sided log-rank test P-value for
EFS was estimated at 0.038. The PFS at 5 years was 75% (63; 83)
versus 93% (83; 97) [HR = 0.3 (0.12; 0.77), P = 0.007] (Figure 3).
Among the 22 patients who failed to respond or who relapsed, 12
patients among the 17 patients in the ABVD arm and 5 patients
among the 5 patients in the BEACOPP arm received stem cell
transplantation after a salvage regimen.
Overall, seven patients died: six in the ABVD arm and one in
the BEACOPP arm. In the ABVD arm, the cause of death was
the HL for three patients, second cancer for two patients and
150 patients randomized
80 in ABVD
70 in BEACOPP
1 Error
2 Patient
Refusal
1 Error
1 Patient
Refusal
77 started treatment
68 started treatment
3 Failure
1 Toxicity
3 Patient
Refusal
2 Other
2 Failure
3 Toxicity
2 Patient
Refusal
68 completed treatment
61 completed treatment
Figure 1. Consort diagram of the study population.
 | Mounier et al.
Volume 25 | No. 8 | August 2014
original articles
Annals of Oncology
arm. There were 20 severe adverse events in the ABVD arm versus
62 in the BEACOPP arm; all were in relation with treatmentrelated morbidity. Grade 3–4 hematological events were reported
during treatment in the ABVD and BEACOPP arms in 8% and
35% of the patients for febrile neutropenia, 42% and 75% of the
patients for granulocytes and 0% and 21% of the patients for platelets. The most frequently reported grade 3 or 4 non-hematological
events in the ABVD and BEACOPP arms were fatigue (2% and
7%) and sensitive neuropathy (0% and 7%). Other grade 3 or 4
adverse events were not clinically relevant in both arms (Table 3).
Second cancers occurred in five ABVD patients and in one
BEACOPP patient (non-HL 2 and 1, lung 1 and 0, digestive 2
and 0). The three lymphoma subtypes were: T-not otherwise
specified and T-angioimmunoblastic for the ABVD cases and
EBV-induced B large cell lymphoma for the BEACOPP case.
The three patients died from their lymphomas. Among the five
ABVD patients, three received second-line HL treatment. The
T-Angioimmunoblastic lymphoma case occurred after DHAP,
and two solid tumors occurred after ASCT following ICE or
DHAP. See supplementary Figure S1 available at Annals of
Oncology online.
accident for the remaining patient. The only death in the
BEACOPP arm was due to a second malignancy. The 5-year OS
was estimated at 92% (81; 97) versus 99% (90; 100) [HR = 0.18
(0.02; 1.53), P = 0.06].
adverse events
A total of 1124 cycles of chemotherapy were administered with
597 administered ABVD and 527 administered BEACOPP. Hospitalizations occurred in 11% (ABVD) and 37% (BEACOPP)
of the patients. The main reason for hospitalization was the administration of chemotherapy (81% of cases). The other reasons
for hospitalization were the management of toxicity (8%) and
tumor symptoms (5%).
The median dose intensity was 98%, 92%, 97% and 98% for
doxorubicin, bleomycin, vinblastine and dacarbazine in the
ABVD arm. It was 100%, 95%, 98%, 100%, 100% and 99% for
doxorubicin, bleomycin, vincristine, etoposide, cyclophosphamide
and procarbazine in the escalated BEACOPP arm and 98%, 90%,
80%, 98%, 99% and 96%, respectively, in the baseline BEACOPP
Table 2. Response after treatment
ABVD
(n = 77)
Response after 8 cycles
Complete response
Unconfirmed complete response
Partial response
Stable disease
Progressive disease
Stop before 8 cycles
n = 68
34
24
5
0
5
n=9
discussion
BEACOPP
(n = 68)
%
50
35
7
0
7
n = 61
28
27
6
0
0
n=7
A homogeneous group of patients with good prognoses with
stage III and IV HL were enrolled in the present phase III randomized trial to assess the range of benefit achieved in increasing
the chemotherapy dose intensity in the conventional range. No
radiotherapy was allowed in this trial. ABVD ×8 cycles was
selected as the standard regimen and BEACOPPescalated ×4 ≥
BEACOPPbaseline ×4 was selected as the experimental arm. The
difference in outcomes was favorable in the experimental arm.
The 5-year PFS was significantly improved in the experimental
%
46
44
10
0
0
Event-free survival* - Primary analysis (ITT population)
with number of subjects at risk
1.0
Survival probability
0.8
0.6
0.4
0.2
0.0
1
2
1: ABVD
2: BEACOPP
80
70
55
54
49
52
45
51
38
40
26
35
18
22
12
12
5
6
0
2
0
0
1
2
3
4
5
6
7
8
9
10
EFS (Years)
Figure 2. Event-free survival according to treatment arm P = 0.07.
Volume 25 | No. 8 | August 2014
doi:10.1093/annonc/mdu189 | 
original articles
Annals of Oncology
Overall survival* (ITT population)
with number of subjects at risk
A
1.0
Survival probability
0.8
0.6
0.4
0.2
0.0
1
2
1: ABVD
2: BEACOPP
80
70
77
66
76
64
58
62
57
50
39
43
25
27
15
15
5
6
0
2
0
0
1
2
3
4
5
6
7
8
9
10
OS (Years)
Progression free survival* (ITT population)
with number of subjects at risk
B
1.0
Survival probability
0.8
0.6
0.4
0.2
0.0
1
2
1: ABVD
2: BEACOPP
80
70
67
63
59
61
53
58
44
46
29
40
18
25
12
13
5
6
0
2
0
0
1
2
3
4
5
6
7
8
9
10
PFS (Years)
Figure 3. (A) Progression-free survival according to treatment arm P = 0.007. (B) Overall survival according to treatment arm P = 0.06.
arm (75% for ABVD versus 93% for BEACOPP, P = 0.008).
However, the significance level was not reached for our primary
end point (5-year EFS 62% for ABVD versus 77% for BEACOPP,
P = 0.07) or for the 5-year OS (92% for ABVD versus 99% for
BEACOPP, P = 0.08).
Good results with ABVD treatment have also been published
by Moccia et al. [8]. In an analysis restricted to patients <65
years old, results from Moccia et al. showed a 5-year OS ranging
from 92% for patients with 2 IPS factors to 98% for patients
with no adverse factors and showed a PFS ranging from 80% to
88%, respectively. In the present trial comparing BEACOPP
with ABVD, treatment was administered full dose with an
 | Mounier et al.
excellent compliance; the lack of significant gain could be
explained by the small sample size of the study population (n =
150). Therefore, although the PFS results revealed the longawaited dose–response effect, it appears that BEACOPP would
be a PFS benefit to 5–7% of the patients if used initially instead
of ABVD. This finding has also been suggested by the two other
studies that randomized ABVD and escalated-BEACOPP in
patients with advanced HL [12, 13]. The baseline IPS (<3 versus
≥3) did not significantly influence the outcome of treatment in
a study conducted by Viviani et al., and the gain in 5-year PFS
was not significant (75% for ABVD versus 90% for BEACOPP,
P = 0.125) in a study conducted by Federico et al.
Volume 25 | No. 8 | August 2014
original articles
Annals of Oncology
the BV–AVD group. At the end of treatment, the CR was estimated at 95% for the patients treated with BV–ABVD and 96%
for BV–AVD.
Table 3. Frequency of grade 3–4 toxicities (all cycles)
Anemia*
Leukopenia*
Neutropenia*
Thrombopenia*
Febrile neutropenia*
Infection without neutropenia
Allergy
Fatigue
Cardiovascular
Digestive
Pulmonary
Peripheral neuropathy
No. of patients
ABVD
%
(n = 77)
BEACOPP
(n = 68)
%
3
6
32
0
6
2
1
2
4
11
1
0
23
55
51
14
24
7
3
5
4
10
1
5
34
81
75
21
35
10
4
7
6
15
1
7
4
8
42
0
8
3
1
3
5
14
1
0
conclusion
* Indicates a P-value <0.05.
The immediate morbidity during treatment was higher in the
BEACOPP arm: 8% of patients experienced febrile neutropenia
in the ABVD arm versus 35% in the BEACOPP arm. However,
all of the complications were manageable with growth factor
support and a strict dose reduction, even in our multicenter
setting, and no deaths occurred due to toxicity [14]. With a
median follow-up period of 5.5 years, late morbidity could not
be assessed. Thus far, no difference has emerged concerning
cardiac and pulmonary functions. However, a striking point was
noted concerning the second solid tumors. Second cancers occurred in five ABVD patients and in one BEACOPP patient
(NHL 2 and 1, lung 1 and 0, other 2 and 0). Among the five
ABVD patients, three received second-line HL treatment. These
findings suggested that HL relapse after ABVD could now be
cured but at the price of complications such as secondary cancers.
At the present time, physicians are not able to predetermine
which patients can be cured by ABVD and which patients will
benefit from escalated BEACOPP. An interim fluorodeoxyglucose PET assessment—i.e. after two cycles of chemotherapy—
can be used to predict a final outcome in HL [15, 16]. These
promising findings lead investigators to launch several ongoing
trials to test PET-based strategies in advanced disease. Another
aim is to improve the prediction of a final outcome using
molecular profiling. Recently, Scott et al. [17] succeeded in
developing a predictor of survival using gene expression measured in a formalin-fixed paraffin-embedded biopsy. There
was a 29% absolute difference in the 5-year OS between the
high- and low-risk groups, independent of the IPS. As the
study population did not receive BEACOPP, further studies are
required to determine whether the high risk of death identified
by the model can be overcome by BEACOPP or by novel
agents combined with ABVD. In a recent trial, Younes et al.
[18] treated 51 patients with six cycles of 0.6–1.2 mg/kg of
brentuximab vedotin plus ABVD (BV–ABVD) or 1.2 mg/kg of
brentuximab vedotin with AVD (BV–AVD). Notably, 11 of the
25 (44%) patients treated with BV–ABVD experienced pulmonary toxicity resulting in two deaths compared with 0% in
Volume 25 | No. 8 | August 2014
As PET and molecular study results are not yet available, treatment with BEACOPP should be a robust option over ABVD
because, even in low-risk patients, it significantly improves PFS.
However, considering the high burden, morbidity and costs
related to BEACOPP, the results of this study raise the issue of
identifying these patients early to be able to offer them an alternative treatment on the ABVD platform.
acknowledgments
We are indebted to the staff of LYSA RC for coordinating the
study, to Christine Stephan for acting as the project leader, to
Samy Bousseta for generating the statistics and to Laurence
Girard for working on the data management. All of the investigators and pathologists (special thank to Jose Audoin) from the
different centers should be recognized for their work as well as
the patients for their participation in the trial. List of active
investigators: M. Divine, PB, P. Fenaux, B. Coiffier, O. Reman,
M. Aoudjhane, A.M. Blaise, D. Bordessoule, JG, A. Bosly,
F. Morschhauser, D. Caillot, H. Gonzalez, P. Lederlin,
R. Bouabdallah, A. Van Hoof, O. Boulat, F. Bauduer,
O. Tournilhac, PC, D. Decaudin, C. Sebban, M. Janvier,
A. Kentos, L. Voillat, JJ, M. Fabbro, PC, J.C. Eisenmann,
C. Martin, B. Christian, A. Ferrant, C. Salanoubat, B. Varet,
K. Bouabdallah, B. De Prijck, X. Levaltier, S. Castaigne,
B. Audhuy, AD, N. Frenkiel, C. Rose, O. Fitoussi, H. Orfeuvre, J.
M. Pignon.
funding
Supported by research grants from French Programme
Hospitalier de Recherche Clinique (PHRC AOM01066 2001)
and Chugai Company.
disclosure
The authors have declared no conflicts of interest.
references
1. Bonadonna G, Viviani S, Bonfante V et al. Survival in Hodgkin’s disease patients—
report of 25 years of experience at the Milan Cancer Institute. Eur J Cancer 2005;
41: 998–1006.
2. Diehl V, Franklin J, Hasenclever D et al. BEACOPP, a new dose-escalated and
accelerated regimen, is at least as effective as COPP/ABVD in patients with
advanced-stage Hodgkin’s lymphoma: interim report from a trial of the German
Hodgkin’s Lymphoma Study Group. J Clin Oncol 1998; 16: 3810–3821.
3. Diehl V, Franklin J, Pfreundschuh M et al. Standard and increased-dose BEACOPP
chemotherapy compared with COPP-ABVD for advanced Hodgkin’s disease. N
Engl J Med 2003; 348: 2386–2395.
4. Engert A, Diehl V, Franklin J et al. Escalated-dose BEACOPP in the treatment of
patients with advanced-stage Hodgkin’s lymphoma: 10 years of follow-up of the
GHSG HD9 study. J Clin Oncol 2009; 27: 4548–4554.
doi:10.1093/annonc/mdu189 | 
original articles
5. Wongso D, Fuchs M, Plutschow A et al. Treatment-related mortality in patients
with advanced-stage Hodgkin lymphoma: an analysis of the German Hodgkin
Study Group. J Clin Oncol 2013; 31: 2819–2824.
6. Skoetz N, Trelle S, Rancea M et al. Effect of initial treatment strategy on survival of
patients with advanced-stage Hodgkin’s lymphoma: a systematic review and
network meta-analysis. Lancet Oncol 2013; 14: 943–952.
7. Hasenclever D, Diehl V. A prognostic score for advanced Hodgkin’s disease:
International Prognostic Factors Project on Advanced Hodgkin’s Disease. N Engl J
Med 1998; 339: 1506–1514.
8. Moccia AA, Donaldson J, Chhanabhai M et al. International Prognostic Score in
advanced-stage Hodgkin’s lymphoma: altered utility in the modern era. J Clin
Oncol 2012; 30: 3383–3388.
9. Carde P, Karrasch M, Fortpied C et al. ABVD (8 cycles) versus BEACOPP (4
escalated cycles 5> 4 baseline) in stage III–IV high-risk Hodgkin lymphoma (HL):
first results of EORTC 20012 Intergroup randomized phase III clinical trial.
American Society of Clinical Oncology 48th annual meeting, June, 1–5 2012. J
Clin Oncol 2012; 30: Abstr 8002.
10. Cheson BD, Horning SJ, Coiffier B et al. Report of an international workshop to
standardize response criteria for non-Hodgkin’s lymphomas. NCI Sponsored
International Working Group. J Clin Oncol 1999; 17: 1244.
11. Hryniuk WM, Goodyear M. The calculation of received dose intensity. J Clin Oncol
1990; 8: 1935–1937.
12. Federico M, Luminari S, Iannitto E et al. ABVD compared with BEACOPP compared
with CEC for the initial treatment of patients with advanced Hodgkin’s lymphoma:
Annals of Oncology
13.
14.
15.
16.
17.
18.
results from the HD2000 Gruppo Italiano per lo Studio dei Linfomi Trial. J Clin
Oncol 2009; 27: 805–811.
Viviani S, Zinzani PL, Rambaldi A et al. ABVD versus BEACOPP for Hodgkin’s
lymphoma when high-dose salvage is planned. N Engl J Med 2011; 365:
203–212.
Borchmann P, Haverkamp H, Diehl V et al. Eight cycles of escalated-dose
BEACOPP compared with four cycles of escalated-dose BEACOPP followed by four
cycles of baseline-dose BEACOPP with or without radiotherapy in patients with
advanced-stage hodgkin’s lymphoma: final analysis of the HD12 trial of the
German Hodgkin Study Group. J Clin Oncol 2011; 29: 4234–4242.
Gallamini A, Hutchings M, Rigacci L et al. Early interim 2-[18F]fluoro-2-deoxy-Dglucose positron emission tomography is prognostically superior to international
prognostic score in advanced-stage Hodgkin’s lymphoma: a report from a joint
Italian–Danish study. J Clin Oncol 2007; 25: 3746–3752.
Engert A, Haverkamp H, Kobe C et al. Reduced-intensity chemotherapy and PETguided radiotherapy in patients with advanced stage Hodgkin’s lymphoma (HD15
trial): a randomised, open-label, phase 3 non-inferiority trial. Lancet 2012; 379:
1791–1799.
Scott DW, Chan FC, Hong F et al. Gene expression-based model using formalinfixed paraffin-embedded biopsies predicts overall survival in advanced-stage
classical Hodgkin lymphoma. J Clin Oncol 2013; 31: 692–700.
Younes A, Connors JM, Park SI et al. Brentuximab vedotin combined with ABVD or
AVD for patients with newly diagnosed Hodgkin’s lymphoma: a phase 1, openlabel, dose-escalation study. Lancet Oncol 2013; 14: 1348–1356.
Annals of Oncology 25: 1628–1635, 2014
doi:10.1093/annonc/mdu173
Published online 14 May 2014
Comprehensive gene expression meta-analysis of head
and neck squamous cell carcinoma microarray data
defines a robust survival predictor
L. De Cecco1, P. Bossi2, L. Locati2, S. Canevari1,3* & L. Licitra2
1
Functional Genomics and Informatics, Department of Experimental Oncology and Molecular Medicine; 2Head and Neck Medical Oncology Unit, Department of Molecular
Oncology; 3Molecular Therapies, Department of Experimental Oncology and Molecular Medicines, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy
Received 2 August 2013; revised 24 February 2014 and 24 April 2014; accepted 28 April 2014
Background: Head and neck squamous cell carcinoma refers to a heterogeneous disease frequently aggressive in its biologic behavior. Despite the improvements in the therapeutic modalities, the long-term survival rate remained unchanged over
the past decade and patients with this type of cancer are at a high risk of developing recurrence. For this reason, there is a great
need to find better ways to foresee outcome, to improve treatment choices, and to enable a more personalized approach.
Patients and methods: Nine microarray gene expression datasets, reporting survival data of a total of 841 samples, were
retrieved from publicly repositories. Three datasets, profiled on the same version of microarray chips, were selected and
merged following a meta-analysis approach to build a training set. The remaining six studies were used as independent validation sets.
Results: The training set led us to identify a 172-gene signature able to stratify patients in low or high risk of relapse [log-rank,
P = 2.44e−05; hazard ratio (HR) = 2.44, 95% confidence interval (CI) 1.58–3.76]. The model based on the 172 genes was validated on the six independent datasets. The performance of the model was challenged against other proposed prognostic signatures (radiosensitivity index, 13-gene oral squamous cell carcinoma signature, hypoxia metagene, 42-gene high-risk
*Correspondence to: Dr Silvana Canevari, Department of Experimental Oncology and
Molecular Medicine, Fondazione IRCCS Istituto Nazionale dei Tumori, via Amadeo 42,
Milan 20133, Italy, Tel: +39-02-23902567; E-mail: [email protected]
© The Author 2014. Published by Oxford University Press on behalf of the European Society for Medical Oncology.
All rights reserved. For permissions, please email: [email protected].