Abstract

Background

Hodgkin Lymphoma (HL) is highly curable when treated accurately. The challenge is to cure patients with the minimal risk of long-term complications. For that, optimal initial diagnostics are required to determine the optimal treatment plan. We offer non-academic hospitals in our Regional Comprehensive Cancer Centre network a centralised review of all diagnostic procedures from patients with newly diagnosed HL. We report our experience on concordances and discrepancies between local findings and central review results.

Patients and methods

A haematologist and radiation oncologist at the Hodgkin Radboud University Nijmegen Medical Centre outpatient clinic examined all patients with newly diagnosed HL between February 2006 and May 2010. In a multidisciplinary lymphoma conference, diagnostic information is reviewed and treatment advice formulated. Discordant findings in pathology, staging and therapy were recorded as ‘minor’, no therapeutic consequences or ‘major’, adapted therapy advice.

Results

Altogether, 125 patients were included. Pathology review showed 86% concordance, with 4% major discordance, mainly nodular lymphocyte predominant sub-type. Revision of initial staging was concordant in 77%; however 15% major discordance of which most were upstaged. This resulted in 19% treatment adaption.

Conclusion

Our findings highlight the discrepancies in interpretation of diagnostic tests. We advocate centralised review process for all newly diagnosed patients with HL.

introduction

Hodgkin lymphoma (HL) is a rare but one of the best curable malignancies. With a median age at diagnosis of 30–35 years, the majority of patients will have a long life expectancy, which could be affected by long-term therapy-related complications. Fifteen to twenty years after treatment second malignancies (e.g. breast or lung cancer) and cardiovascular diseases overtake the risk of mortality due to treatment-refractory HL [1–4]. The main challenge faced by physicians treating patients with HL is to cure their patients with the minimal risk for long-term complications [5]. Standard chemotherapy has switched from the mitoxin, vincristin, procarbazin and prednisone (MOPP) schedule or MOPP-like variants to adriamycin, bleomycin, vinblastin and dacarbazin (ABVD) or its variants, thereby reducing the risk of secondary leukaemias and infertility [6–8]. ABVD may induce anthracycline-related cardiotoxicity and bleomycine-induced lung damage. On the other hand, the successful introduction of bleomycin, etoposide, adriamycin, cyclophosphamide, vincristin, procarbazin and prednisone (BEACOPP) escalated indeed increased efficacy in terms of progression-free and overall survival for patients with advanced disease, but at the cost of the increased toxicity as seen in the historical MOPP/MOPP-like schedules [9, 10].

Reduction of the total number of required cycles of chemotherapy may decrease the risk of toxicity but cure rates should not come down. Reduction, or even omittance of radiotherapy, will certainly abolish certain serious late effects such as infield malignancies (breast and lung cancer) or premature cardiovascular disease, but has to be balanced against the potential higher risk of relapse. In advanced disease, radiotherapy (RT) can be restricted to 15%–20% of patients after BEACOPP escalated, using very careful post-chemotherapy restaging including 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) scan [11]. The European Organisation for Research and Treatment of Cancer (EORTC) Lymphoma group is pioneering the involved node RT principle in the combined modality treatment in early-stage disease, to further reduce the radiation fields [12]. Evidently, these newer approaches ask for a meticulous staging procedure, reliable and reproducible identification of involved nodes and extranodal sites and a precise classification in the various predefined risk groups (Ann Arbor stage, EORTC stage I/II risk criteria and the International Prognostic Score for advanced stage disease). Preferably, this is done by experienced specialists in clinical haematology, radiation oncology, pathology, radiology and nuclear medicine [11, 13, 14]. Therefore, it seems rational to have a centralised review of the diagnostic procedures, because it is known that the reliability of pathology and imaging is not optimal in centres with a low volume of specific forms of cancer.

In The Netherlands, patients with HL are treated in university medical centres as well as in community hospitals. Community hospitals diagnose and treat between 1 and 10 new patients with HL yearly. Although the consultation of specialist expertise is well organised within the framework of our Dutch Comprehensive Cancer Centre consultation network, the given advice is based upon written or vocal information often without a structured review of all available diagnostic data. Therefore, we decided to offer our affiliated community hospitals, a centralised review for all newly diagnosed patients with HL to optimise diagnosis and treatment while maintaining the infrastructure of having patients being treated in their immediate vicinity. In close cooperation with the participating regional hospitals started in 2006, our HL joint outpatient clinic were all new patients with HL were seen by one haematologist and one radiation oncologist of the Radboud University Nijmegen Medical Centre (RUN MC). In this report, we review our experience on the first 125 patients focussing on the concordances and discrepancies we have observed between local findings and the specialised opinion.

methods

All newly diagnosed, previously untreated patients with HL who visited the Hodgkin outpatient clinic from January 2006 to May 2010 were evaluated. The regional network consists of the following community hospitals: Rijnstate Hospital (Arnhem), Canisius Wilhelmina Hospital (Nijmegen), Hospital De Gelderse Vallei (Ede), Slingeland Hospital (Doetinchem), VieCuri Medical Centre Noord-Limburg (Venlo), St. Jansdal (Harderwijk) and Streekziekenhuis (Zevenaar).

Central revision was performed on histo- and immunopathology (with additional molecular techniques when needed) and imaging (radiology and nuclear medicine) from the referring hospitals. Relevant clinical data, from the referring hospitals and the RUN MC were collected from the article and electronic patient records. For this study, all the referring hospitals were visited for the verification of all collected data and the completion of follow-up data. The patients were seen once, within a week after referral from the community hospital, on the joint HL outpatient clinic by a team consisting of a haematologist and a radiation oncologist, for medical history and physical examination. Two days after the HL outpatient clinic visit all available data were centrally reviewed during the weekly multidisciplinary conference attended by haematologists, radiation oncologists, pathologists, radiologists and nuclear medicine physicians. The final diagnosis, stage attribution and treatment advice were defined and the same day communicated to the referring physician.

The conclusions on pathology, staging and treatment advice were categorised as either concordant or discordant with that of the referring hospital. Concordance implies that the RUN MC experts agreed with the clinician, pathologist, radiologist and nuclear medicine physician of the referring hospital. Discordance means that there was a difference of opinion between the referring specialist and the RUN MC. The discordances were categorised as minor, when the difference did not change the treatment advice, and major when the change in the pathology or staging resulted in a treatment adjustment.

Statistical analysis was performed using SPSS 16.0.01, release 16.0.2, 10-april-2008. Descriptive techniques were used for specifying the population and observed discordances.

results

A total of 125 patients visited the HL outpatient clinic between January 2006 and May 2010. Their clinical data are summarized in Table 1. The median age at first visit was 37 years (range 16–83) and 52% were male. The median age and male/female distribution is as expected.

Table 1.

Patient characteristics

 Value Normal value 
Median age, years (range) 37 (16–83)  
Sex 
 Male 65 (52)  
 Female 60 (48)  
Referring hospital 
 Canisius Wilhelmina Hospital, Nijmegen, n (%) 34 (27)  
 Rijnstate Hospital, Arnhem, n (%) 24 (19)  
 Hospital De Gelderse Vallei, Ede, n (%) 24 (19)  
 Slingeland Hospital, Doetinchem, n (%) 16 (13)  
 VieCuri Medical Centre Noord-Limburg, Venlo, n (%) 16 (13)  
 St. Jansdal, Harderwijk, n (%) 6 (5)  
 Streekziekenhuis, Zevenaar, n (%) 5 (4)  
Laboratory results   
 Mean ESR, mm/h (range) (n = 124) 41 (2–138) <20 
 Mean haemoglobin, mmol/L (range) (n = 124) 8.1 (4.2–11.0) Male: 8.1–10.7 
Female: 7.3–9.7 
 Mean leukocyte count, ×109 (range) (n = 124) 9.3 (1.5–21.0) 3.5–11 
Mean lymphocyte count, ×109 (range) (n = 116) 1.6 (0.2–3.7) 2–4 
Mean albumin, g/L (range) (n = 112) 41 (23–51) 35–50 
 Value Normal value 
Median age, years (range) 37 (16–83)  
Sex 
 Male 65 (52)  
 Female 60 (48)  
Referring hospital 
 Canisius Wilhelmina Hospital, Nijmegen, n (%) 34 (27)  
 Rijnstate Hospital, Arnhem, n (%) 24 (19)  
 Hospital De Gelderse Vallei, Ede, n (%) 24 (19)  
 Slingeland Hospital, Doetinchem, n (%) 16 (13)  
 VieCuri Medical Centre Noord-Limburg, Venlo, n (%) 16 (13)  
 St. Jansdal, Harderwijk, n (%) 6 (5)  
 Streekziekenhuis, Zevenaar, n (%) 5 (4)  
Laboratory results   
 Mean ESR, mm/h (range) (n = 124) 41 (2–138) <20 
 Mean haemoglobin, mmol/L (range) (n = 124) 8.1 (4.2–11.0) Male: 8.1–10.7 
Female: 7.3–9.7 
 Mean leukocyte count, ×109 (range) (n = 124) 9.3 (1.5–21.0) 3.5–11 
Mean lymphocyte count, ×109 (range) (n = 116) 1.6 (0.2–3.7) 2–4 
Mean albumin, g/L (range) (n = 112) 41 (23–51) 35–50 

pathology

Agreement between the pathologists from the referring hospital and the RUN MC was reached in 108/125 cases (86%) (Table 2). Minor discordances in the sub-typing of HL concerned a change between the mixed cellularity (MCcHL) and nodular sclerosing (NScHL) sub-type (n = 5) and between not otherwise specified (NOS) and MCcHL or NScHL (n = 7) (Table 2).

Table 2.

Comparison in histopathology between the referring hospital and centralised revision

  Histopathology referring hospital
 
  NScHL MCcHL LRcHL NLPHL NOS 
 NScHL 75   
 MCcHL 10   
Histopathology centralised revision LRcHL   5  
 NLPHL   10  
 NOS   8 
 Others   1a 1b  
  Histopathology referring hospital
 
  NScHL MCcHL LRcHL NLPHL NOS 
 NScHL 75   
 MCcHL 10   
Histopathology centralised revision LRcHL   5  
 NLPHL   10  
 NOS   8 
 Others   1a 1b  

aDiffuse large B-cell lymphoma.

bEBV positive lymphoproliferative Hodgkin-like lymphoproliferation.

NScHL, nodular sclerosis; MCcHL, mixed cellularity; LRcHL, lymphocyte rich; NLPHL, nodular lymphocyte predominant; NOS, not otherwise specified.

Major discordance was established in 5/125 cases (4%). All of these were lymphocyte rich lesions and the well-known difficult differential diagnosis between lymphocyte-rich classical HL (LRcHL), nodular lymphocyte predominant HL (NLPHL) and diffuse large B-cell NHL (DLBCL) was the problem in four cases. In one case, the diagnosis was changed from LRcHL to methotrexate (MTX)-related Epstein–Barr positive Hodgkin-like lymphoproliferation due to the clinical presentation.

staging

The Ann Arbor stage could be attributed to 123/125 (98%) patients at the central review. In one patient, FDG-PET-CT scan could not be timely retrieved and, in one, the diagnosis changed to DLBCL (Table 3). There were 113 FDG-PET scans with a diagnostic CT scan reviewed and 11 FDG-PET scans with a low-dose CT scan.

Table 3.

Differences in staging between the referring hospital and centralised revision

  Ann Arbor referring hospital
 
  Stage I favourable Stage I unfavourable Stage II favourable Stage II unfavourable Stage III good risk Stage III poor risk Stage IV good risk Stage IV poor risk 
 Stage I favourable 9       
 Stage I unfavourable  4      
 Stage II favourable  21    
Ann Arbor centralised revision Stage II unfavourable  26     
Stage III good risk   14    
 Stage III poor risk     6   
 Stage IV good risk    5  
 Stage IV poor risk      10 
 Missing data/other     1a  1b  
  Ann Arbor referring hospital
 
  Stage I favourable Stage I unfavourable Stage II favourable Stage II unfavourable Stage III good risk Stage III poor risk Stage IV good risk Stage IV poor risk 
 Stage I favourable 9       
 Stage I unfavourable  4      
 Stage II favourable  21    
Ann Arbor centralised revision Stage II unfavourable  26     
Stage III good risk   14    
 Stage III poor risk     6   
 Stage IV good risk    5  
 Stage IV poor risk      10 
 Missing data/other     1a  1b  

aCT and FDG-PET scan could not be timely retrieved.

bDLBCL.

Of the 123 cases, which could be compared, 95 (77%) were concordant (Table 3). A total of 28 patients (23%) had a discordant stage after revision, 10 were minor discordant and in 18 a major discordance was revealed (Table 3). Details of the 18 major discordant cases include:

  • Downscaling after centralised revision

    • Clinical stage (CS) III to stage CS II due to negative FDG-PET scan on doubtful infra-diaphragmatic lymph node n = 1

    • CS II unfavourable to CS II favourable due to negative FDG-PET scan on one suspected lymph node area n = 1

  • Upscaling after centralised revision

    • CS I/II to CS III/IV due to FDG-PET driven revision of CT scan with suspected infra-diaphragmatic nodes n = 8

    • CS I/II favourable to CS I/II unfavourable due to abnormal nodes on FDG-PET scan correlating with revised CT scan findings n = 3

    • CS I/II favourable to CS I/II unfavourable due to elevated ESR and/or B symptoms n = 3

    • CS II favourable to CS IV good risk due to bone lesions on revised FDG-PET scan n = 1

    • CS III to CS IV because of suspected liver lesion on revised FDG-PET scan n = 1

therapy

In 124/125 patients, a central treatment advice could be given. In the group of 124 patients, the referral hospital had already specified the treatment advice in 104 while, in 20 patients, the treatment was not yet defined.

The treatment advice was concordant in 84/104 (81%) cases. In 20 patients (19%), the central revision process led to a change in treatment advice based on changes in the pathology and/or staging results (Table 4).

Table 4.

Differences in treatment proposal between the referring hospital and centralised revision

  Treatment proposal referring hospital
 
  IF-RT ABVD × 6 ABVD × 8 ChlVPP × 6-8 Other chemo ABVD × 3 + IN-RT ABVD × 4 + IN-RT Missing data 
 IF-RT 8       
 ABDV × 6  27    
 ABVD × 8   1      
 ChlVPP × 6-8    1    
 Other chemo    2   
Treatment proposal centralised revision ABVD × 3 + IN-RT    22 
ABVD × 4 + IN-RT      23 
 Missing data        
 Other chemo + RT      1a 1b 
 Other treatment       2c 
  Treatment proposal referring hospital
 
  IF-RT ABVD × 6 ABVD × 8 ChlVPP × 6-8 Other chemo ABVD × 3 + IN-RT ABVD × 4 + IN-RT Missing data 
 IF-RT 8       
 ABDV × 6  27    
 ABVD × 8   1      
 ChlVPP × 6-8    1    
 Other chemo    2   
Treatment proposal centralised revision ABVD × 3 + IN-RT    22 
ABVD × 4 + IN-RT      23 
 Missing data        
 Other chemo + RT      1a 1b 
 Other treatment       2c 

a3 × MOPP with IN-RT.

bChlVPP with IN-RT.

cWatchful waiting.

Changes in RT fields occurred in four patients:

  • Involved field radiotherapy (IF-RT) changed to ABVD + IN-RT due to change in diagnosis from NLPHL to LRcHL n = 1 or CS I favourable to CS II favourable n = 1

  • ABVD × 3 + involved node radiotherapy (IN-RT) changed to IF-RT due to change in diagnosis from LRcHL to NLPHL n = 1

  • IF-RT changed other chemotherapy due to change in histological diagnosis from NLPHL to DLBCL n = 1

Combined changes in the number of ABVD cycles and involved nodes for the IN-RT in 12 patients:

  • CS II favourable changed to CS II unfavourable resulting in ABVD × 4 instead of × 3 + IN-RT n = 5

  • CS II unfavourable changed to CS II favourable resulting in ABVD × 3 instead of × 4 + IN-RT n = 1

  • CS I/II change to CS III/IV resulted in change from ABVD × 3/4 + IN-RT to ABVD × 6/8 n = 5

  • CS III/IV change to CS I/II resulting in change from ABVD × 6 to ABVD × 3 + IN-RT n = 1

Other changes in four patients:

  • Change type of chemotherapy ABVD to other schedule due to co-morbidity n = 2

  • Change ABVD + IN-RT to stop MTX in MTX related Hodgkin-like lymphoma n = 1

  • Change ABVD + IN-RT in close observation in unconfirmed HL diagnosis n = 1

discussion

In this report, we evaluated our experience with a centralised diagnostic review process for 125 previously untreated patients with HL, initially diagnosed in one of the cooperating neighbouring community hospitals. Our results indicate that this representative group of patients with HL a 20% change in treatment advice was given upon revision of the diagnostic procedures.

In the past, some series have addressed the significance of a centralised pathology panel or FDG-PET scan panel [15–22]. This is the first series to combine a review of pathology, imaging and treatment advice in one process. Our results indicate that an experienced panel for diagnosis, imaging and treatment advice is needed because treatment is increasingly tailored to individual patient characteristics including meticulous staging, to prevent under- or over-treatment.

The inter-observer concordance in pathology of HL varies between 49% and 96% [15–18, 23]. We observed a high concordant pathology rate of 86% in our central review. The high agreement percentage is probably attributable to the regional lymphoma panel, in which all newly diagnosed malignant lymphomas are centrally reviewed. Evaluation of this panel is in progress, but preliminary data indicate that about 50% of all new lymphomas are actually referred to the panel. Recently, the surplus value of central review was highlighted by Proctor et al. [23]. In our review, only 4% of cases had major discordant histopathology leading to a treatment adjustment. The majority of this major discordant histopathology was due to NLPHL changed into LRcHL or DLBCL or vice versa a well-known pitfall [24, 25]. In fact, LRcHL was only relatively recently identified among NLPHL cases which were centrally reviewed in the framework of a clinical trial. The European task force on lymphoma project reviewed 388 samples of NLPHL; 56% were concordant and of the discordant 30% was changed into newly created subgroup LRcHL [24]. We only changed 8% of the NLPHL into LRcHL, since this category is now well recognised in the WHO 2008 classification. The importance of correct clinical data when pathology is performed was illustrated by a case that was morphologically and phenotypically HL, but appeared to be a MTX-related Epstein–Barr positive Hodgkin-like lymphoproliferation. This is a well-known pitfall which can only be avoided by proper clinical information. Obviously, this is a very relevant change, since this patient does not need treatment for HL.

The reduction of radiation fields from involved field to involved node enhances the risk of under treatment if the involved nodes are not correctly determined [12]. As most relapses occur in the initially involved lymph node(s) in patients with early-stage HL treated with only chemotherapy, accurate staging is indispensable [26–29]. The increasing use of FDG-PET scanning not only in response to therapy evaluation but also in primary staging, asks for standardization of interpretation of the scans. There is less discordance between experienced nuclear physicians than between less experienced and expert readers [19, 20, 22]. Pre-therapy FDG-PET scan is not mandatory but strongly advised and widely used for patients with HL [30]. Zijlstra et al. identified a higher than anticipated inter-observer disagreement, only 56% of the baseline scans being concordant with the expert opinion [19]. There was a difference between the FDG-PET positive scans and FDG-PET negative scans at baseline. The FDG-PET positive baseline scans were concordant in 82%, and the FDG-PET negative baseline scans were only concordant in 45% [19]. Hofman et al. observed a high intra- and inter-observer agreement of the overall Ann Arbor stages of 0.71–0.91 with a 95% confidence interval [20]. This high concordance is probably due to the fact that their observers were experienced reviewers of FDG-PET-CT scans. In our series, we observe a concordance of 77%, which is in agreement to the literature. Most of our discordance was due to the fact that the nodes or lesions were missed by the first nuclear medicine physicians (n = 16), other reasons were a very low threshold for positive nodes (n = 6) which let to downscaling and upscaling after correlating the FDG-PET scan with the CT scan (n = 3). Two-thirds of the discordance was in patients who were seen during the first 2 years of our HL out-patients clinic. The reason the concordance is not as high as mentioned in Hofman et al. is due to the fact that we have a greater variation in inter-observer FDG-PET scan experience. Standardization of interpretation methods and improvement of the technology results in better concordance than mentioned by Zijlstra et al. This emphasises the fact that nuclear medical physicians need learning sessions and that interpretation should be standardized.

RT fields were changed in our review in 10/104 patients with early-stage HL based on the centralised expert review of the imaging results. It is self-evident that a change from an early stage to an advanced stage as encountered in 9.6% of our patients has even more clinical impact. The literature emphases that more experienced interpreters have fewer false-positives results and that standardization of FDG-PET-scan protocols is necessary [19, 20].

The centralised review process on pathology and staging resulted in an overall change in treatment advice for 19% of newly diagnosed patients with HL in our region. Whether this adaptation of treatment actually improves the outcome of the patients cannot be answered by this prospective observational study. However, our findings highlight the heterogeneity in interpretation of diagnostic tests. As modern treatment approaches rely on correct initial diagnosis and staging using modern pathologic classification and imaging techniques, we strongly advocate a centralised review process for all newly diagnosed patients with such a rare disease as HL.

disclosure

The authors have declared no conflicts of interest.

references

1
Czuczman
M
Straus
D
Gribben
J
, et al.  . 
Management options, survivorship, and emerging treatment strategies for follicular and Hodgkin lymphomas
Leuk Lymphoma
 , 
2010
, vol. 
51
 (pg. 
41
-
49
)
2
Hodgson
DC
Grunfeld
E
Gunraj
N
, et al.  . 
A population-based study of follow-up care for Hodgkin lymphoma survivors: opportunities to improve surveillance for relapse and late effects
Cancer
 , 
2010
, vol. 
116
 (pg. 
3417
-
3425
)
3
van den Belt-Dusebout
AW
Aleman
BM
Gietema
JA
, et al.  . 
[Long-term complications following treatment of testicular cancer and Hodgkin lymphoma]
Ned Tijdschr Geneeskd
 , 
2010
, vol. 
154
 pg. 
A2229
 
4
Ng
AK
Review of the cardiac long-term effects of therapy for Hodgkin lymphoma
Br J Haematol
 , 
2011
, vol. 
154
 (pg. 
23
-
31
)
5
Engert
A
Plutschow
A
Eich
HT
, et al.  . 
Reduced treatment intensity in patients with early-stage Hodgkin's lymphoma
N Engl J Med
 , 
2010
, vol. 
363
 (pg. 
640
-
652
)
6
Santoro
A
Bonadonna
G
Valagussa
P
, et al.  . 
Long-term results of combined chemotherapy-radiotherapy approach in Hodgkin's disease: superiority of ABVD plus radiotherapy versus MOPP plus radiotherapy
J Clin Oncol
 , 
1987
, vol. 
5
 (pg. 
27
-
37
)
7
Carde
P
Hagenbeek
A
Hayat
M
, et al.  . 
Clinical staging versus laparotomy and combined modality with MOPP versus ABVD in early-stage Hodgkin's disease: the H6 twin randomized trials from the European Organization for Research and Treatment of Cancer Lymphoma Cooperative Group
J Clin Oncol
 , 
1993
, vol. 
11
 (pg. 
2258
-
2272
)
8
Anselmo
AP
Cartoni
C
Bellantuono
P
, et al.  . 
Risk of infertility in patients with Hodgkin's disease treated with ABVD vs MOPP vs ABVD/MOPP
Haematologica
 , 
1990
, vol. 
75
 (pg. 
155
-
158
)
9
Sieber
M
Engert
A
Diehl
V
Treatment of Hodgkin's disease: results and current concepts of the German Hodgkin's Lymphoma Study Group
Ann Oncol
 , 
2000
, vol. 
11
 
Suppl 1
(pg. 
81
-
85
)
10
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
, vol. 
348
 (pg. 
2386
-
2395
)
11
Gallamini
A
Positron emission tomography scanning: a new paradigm for the management of Hodgkin's lymphoma
Haematologica
 , 
2010
, vol. 
95
 (pg. 
1046
-
1048
)
12
Girinsky
T
Van Der
MR
Specht
L
, et al.  . 
Involved-node radiotherapy (INRT) in patients with early Hodgkin lymphoma: concepts and guidelines
Radiother Oncol
 , 
2006
, vol. 
79
 (pg. 
270
-
277
)
13
Lister
TA
Crowther
D
Sutcliffe
SB
, et al.  . 
Report of a committee convened to discuss the evaluation and staging of patients with Hodgkin's disease: Cotswolds meeting
J Clin Oncol
 , 
1989
, vol. 
7
 (pg. 
1630
-
1636
)
14
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
, vol. 
339
 (pg. 
1506
-
1514
)
15
Glaser
SL
Dorfman
RF
Clarke
CA
Expert review of the diagnosis and histologic classification of Hodgkin disease in a population-based cancer registry: interobserver reliability and impact on incidence and survival rates
Cancer
 , 
2001
, vol. 
92
 (pg. 
218
-
224
)
16
Georgii
A
Fischer
R
Hubner
K
, et al.  . 
Classification of Hodgkin's disease biopsies by a panel of four histopathologists. Report of 1,140 patients from the German National Trial
Leuk Lymphoma
 , 
1993
, vol. 
9
 (pg. 
365
-
370
)
17
Velez-Garcia
E
Durant
J
Gams
R
, et al.  . 
Results of a uniform histopathologic review system of lymphoma cases. A ten-year study from the Southeastern Cancer Study Group
Cancer
 , 
1983
, vol. 
52
 (pg. 
675
-
679
)
18
Woodruff
RD
Reviewing histologic diagnosis of lymphoma: comparison of original and review diagnoses in 269 cases
Arch Pathol Lab Med
 , 
1981
, vol. 
105
 (pg. 
573
-
576
)
19
Zijlstra
JM
Comans
EF
van
LA
, et al.  . 
FDG PET in lymphoma: the need for standardization of interpretation. An observer variation study
Nucl Med Commun
 , 
2007
, vol. 
28
 (pg. 
798
-
803
)
20
Hofman
MS
Smeeton
NC
Rankin
SC
, et al.  . 
Observer variation in interpreting 18F-FDG PET/CT findings for lymphoma staging
J Nucl Med
 , 
2009
, vol. 
50
 (pg. 
1594
-
1597
)
21
Delbeke
D
Stroobants
S
de
KE
, et al.  . 
Expert opinions on positron emission tomography and computed tomography imaging in lymphoma
Oncologist
 , 
2009
, vol. 
14
 
Suppl 2
(pg. 
30
-
40
)
22
Barrington
SF
Qian
W
Somer
EJ
, et al.  . 
Concordance between four European centres of PET reporting criteria designed for use in multicentre trials in Hodgkin lymphoma
Eur J Nucl Med Mol Imaging
 , 
2010
, vol. 
37
 (pg. 
1824
-
1833
)
23
Proctor
IE
McNamara
C
Rodriguez-Justo
M
, et al.  . 
Importance of expert central review in the diagnosis of lymphoid malignancies in a regional cancer network
J Clin Oncol
 , 
2011
, vol. 
29
 (pg. 
1431
-
1435
)
24
Anagnostopoulos
I
Hansmann
ML
Franssila
K
, et al.  . 
European Task Force on Lymphoma project on lymphocyte predominance Hodgkin disease: histologic and immunohistologic analysis of submitted cases reveals 2 types of Hodgkin disease with a nodular growth pattern and abundant lymphocytes
Blood
 , 
2000
, vol. 
96
 (pg. 
1889
-
1899
)
25
Boudova
L
Torlakovic
E
Delabie
J
, et al.  . 
Nodular lymphocyte-predominant Hodgkin lymphoma with nodules resembling T-cell/histiocyte-rich B-cell lymphoma: differential diagnosis between nodular lymphocyte-predominant Hodgkin lymphoma and T-cell/histiocyte-rich B-cell lymphoma
Blood
 , 
2003
, vol. 
102
 (pg. 
3753
-
3758
)
26
Longo
DL
Glatstein
E
Duffey
PL
, et al.  . 
Radiation therapy versus combination chemotherapy in the treatment of early-stage Hodgkin's disease: seven-year results of a prospective randomized trial
J Clin Oncol
 , 
1991
, vol. 
9
 (pg. 
906
-
917
)
27
Biti
GP
Cimino
G
Cartoni
C
, et al.  . 
Extended-field radiotherapy is superior to MOPP chemotherapy for the treatment of pathologic stage I-IIA Hodgkin's disease: eight-year update of an Italian prospective randomized study
J Clin Oncol
 , 
1992
, vol. 
10
 (pg. 
378
-
382
)
28
Straus
DJ
Portlock
CS
Qin
J
, et al.  . 
Results of a prospective randomized clinical trial of doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) followed by radiation therapy (RT) versus ABVD alone for stages I, II, and IIIA nonbulky Hodgkin disease
Blood
 , 
2004
, vol. 
104
 (pg. 
3483
-
3489
)
29
Shahidi
M
Kamangari
N
Ashley
S
, et al.  . 
Site of relapse after chemotherapy alone for stage I and II Hodgkin's disease
Radiother Oncol
 , 
2006
, vol. 
78
 (pg. 
1
-
5
)
30
Seam
P
Juweid
ME
Cheson
BD
The role of FDG-PET scans in patients with lymphoma
Blood
 , 
2007
, vol. 
110
 (pg. 
3507
-
3516
)