G r ay Z one Lymp h o ma

Current Diagnosis and Treatment

Options

Monika Pilichowska, MDa, Athena Kritharis, MD

b
,
Andrew M. Evens, DO, MScb,*

KEYWORDS
 Gray zone lymphoma  B-cell  Classical Hodgkin lymphoma
 Diffuse large B-cell lymphoma  Biology  Prognosis  Treatment

KEY POINTS
 The WHO recognizes a distinct category of B-cell lymphoma, unclassifiable, with features
intermediate between diffuse large B-cell lymphoma (DLBCL) and classical Hodgkin lym-
phoma (cHL), also known as gray zone lymphoma (GZL).
 The immunophenotype of GZL is frequently discordant with tumor cells morphologically
resembling DLBCL, but immunophenotypically being more consistent with cHL, and
vice versa.
 Clinically, the initial case descriptions of GZL were primarily with mediastinal presentation;
however, a nonmediastinal (systemic) clinical subtype is now fully recognized.
 Regardless of clinical presentation, patients with GZL have relatively high relapse rates,
especially compared with primary mediastinal DLBCL and cHL.
 Off of a clinical trial, we advocate treating GZL with a DLBCL-directed regimen (eg,
R-CHOP or DA-EPOCH-R) with consideration for consolidative radiotherapy for bulk
disease.

INTRODUCTION

Gray zone lymphoma (GZL) is a rare neoplasm initially described in 2005 based on the
recognition of an aggressive subset of primary mediastinal B-cell lymphoma (PMBCL)
with poor prognosis.1 GZL was officially first recognized in the World Health Organiza-
tion classification as a distinct entity in 2008 as a B-cell lymphoma, unclassifiable, with
features intermediate between diffuse large B-cell lymphoma (DLBCL) and classical
Hodgkin lymphoma (cHL).2 Since this time, there has been more insight into the

a
Department of Pathology and Laboratory Medicine, Tufts Medical Center, 800 Washington
Street, Boston, MA 02111, USA; b Division of Hematology/Oncology, Tufts Medical Center,
800 Washington Street, Boston, MA 02111, USA
* Corresponding author.
E-mail address: AEvens@tuftsmedicalcenter.org

Hematol Oncol Clin N Am 30 (2016) 1251–1260

http://dx.doi.org/10.1016/j.hoc.2016.07.006 hemonc.theclinics.com
0889-8588/16/ª 2016 Elsevier Inc. All rights reserved.
1252 Pilichowska et al

pathologic diagnosis, biology, and clinical outcomes for patients with GZL. Although
initially clinically described as primarily involving the mediastinum,1,3,4 further analyses
have elucidated primary mediastinal and nonmediastinal (systemic) clinical disease
presentations.5,6 Notably, this article does not include discussion of the gray zone en-
tity of lymphoma with features of intermediate DLBCL and Burkitt lymphoma.
GZL with mediastinal presentation (MGZL) is similar in presentation to cHL in part as
it typically occurs in young adults in the third and fourth decade. Conversely, patients
with GZL without mediastinal presentation (NMGZL) are typically older and more often
present with advanced-stage disease. Furthermore, unlike cHL and PMBCL, both
MGZL and NMGZL have a more aggressive clinical course with comparatively inferior
outcomes. Given its newer recognition and relative paucity of associated clinical data,
continued description of this unique disease entity is warranted. The following is a
detailed review of the current literature and descriptions of the diagnosis, biology,
prognosis, and treatment of GZL.

CASE STUDY

A 26-year-old Middle Eastern man was seen for second opinion of a new diagnosis of
stage IIBXE cHL. On initial presentation, he experienced several weeks of nonproduc-
tive cough and drenching night sweats. A chest radiograph showed left upper lobe
consolidation and he was started on antibiotics for a presumed pneumonia. His symp-
toms did not abate. Computerized tomography of the chest indicated multiple soft tis-
sue masses that filled the upper mediastinum including an 11 cm  14 cm anterior
mediastinal mass with hilar adenopathy. Bronchoscopy at outside hospital with left
anterior mediastinoscopy was performed. Based on morphology and positivity for
CD15 and CD30, an initial diagnosis of nodular sclerosis Hodgkin lymphoma was
made. The laboratory data were notable for a mild anemia, with hemoglobin
12 g/dL and hematocrit of 36% without further abnormalities. There was no evidence
of bone marrow involvement on staging biopsy.
Additional pathologic review conducted at our institution and the National Institutes
of Health (NIH) noted unusual morphology and immunophenotype. In addition to areas
of cHL, the tumor exhibited extensive syncytial sheets of large cells, Reed-Sternberg
cells, and variants. The neoplastic cells showed strong positivity for CD20 and CD15
with strong to variable positivity for CD30 (Fig. 1). The final diagnosis was B-cell lym-
phoma, unclassifiable, with features intermediate between DLBCL and cHL (ie, GZL).
The patient was treated with cyclophosphamide, doxorubicin, oncovin, and predni-
sone and rituximab (CHOP-R) for six cycles. The patient had a Deauville score of 3 af-
ter two cycles and was Deauville score 2 at end of chemotherapy; he subsequently
received mediastinal involved-node radiation therapy (3000 cGy). The patient currently
remains disease-free 34 months following initial diagnosis.

BIOLOGY

GZL is known to arise from an altered B cell. In GZL, the putative cell of origin is thymic
B cell, which is also a common precursor for cHL and PMBCL. This common precur-
sor hypothesis explains why cHL, PMBCL, or GZL may relapse as the other related
entity and why they are found as synchronous neoplasms.7 The notion of a common
cell of origin is supported by data examining gene expression signatures of GZL,
PMBCL, and cHL being similar as were numerous genetic abnormalities, such as
gains in chromosome 9p and 2p.8 However, because GZL may differentiate in either
direction (eg, Reed-Sternberg cell/variant or DLBCL), epigenetic changes as opposed
 Gray Zone Lymphoma 1253

Fig. 1. Gray zone lymphoma: World Health Organization B-cell lymphoma, unclassifiable,
with features intermediate between DLBCL and cHL. A 26-year-old man presenting with
bulky mediastinal mass. (A) The tumor is composed of large pleomorphic cells, some resem-
bling Reed-Sternberg cells and variants occurring in sheets as seen in DLBCL (hematoxylin-
eosin, original magnification 200). (B) In some areas, the morphology resembles cHL
with infrequent Reed-Sternberg cells occurring on inflammatory background with small
lymphocytes, histiocytes, and eosinophils (hematoxylin-eosin, original magnification
200). The neoplastic cells in DLBCL-like and cHL-like areas are strongly positive for CD20
(C), CD15 (D), and CD30 (E) (immunoperoxidase, original magnification 200).

to genetic aberrations may more strongly influence the final morphology and
immunophenotype.
Eberle and colleagues9 performed a DNA methylation analyses of 30 cases
including cHL, PMBCL, and GZL. They demonstrated a close relationship between
the three diseases but with unique signatures that distinguished GZL from HL and
PMBCL. These included lack of de novo hypomethylation in cHL, hypomethylation
of HOXA5 in GZL, and hypermethylation of EPHA7 and DAPK1 in PMBCL. In addition,
the nuclear factor-kB pathway was highly enriched and was important to the patho-
genesis of all three disease entities. Notably, most of the aforementioned biologic
studies have been conducted among patients with MGZL; it is not known if NMGZL
has similar or distinct biology.

DIAGNOSIS

The diagnosis of GZL should be made by expert pathologic evaluation of the involved
tissue preferably obtained by an excisional or incisional biopsy. GZL is highly variable
with each tumor displaying a unique set of features. A spectrum of morphologies
known with cHL and DLBCL can occur and divergent morphologic areas are seen
within the same tumor necessitating extensive sampling for the correct diagnosis.
1254 Pilichowska et al

The neoplastic cells are usually large with centroblastic or immunoblastic appearance,
high degree of pleomorphism, and can include bona fide Reed-Sternberg cells and
variants. Mummified cells can be present. In cases resembling PMBCL the neoplastic
cells occur in sheets. In cases resembling cHL the neoplastic cells are sparse and
occur on mixed inflammatory background with small lymphocytes, histiocytes, and
eosinophils. These cases can exhibit significant degree of fibrosis and distinction
from true cHL is difficult. Necrosis can be present and be extensive. Neutrophilic in-
filtrates, however, are not a usual feature.
Similar to morphologic findings, the immunophenotype of GZL is variable and has
transitional features between cHL and DLBCL (Table 1). Tumors resembling cHL
can show prominent CD20, weaker/absent CD30, and absent CD15, whereas tumors
resembling DLBCL may be strongly positive for CD30 and CD15 with negative CD20
and CD79a. Because the B-cell program is preserved, B-cell transcription factors,
such as PAX5, OCT2, and BOB1, are positive in neoplastic cells.7 In comparison
with a cohort of 51 patients with PMBCL, patients with MGZL were more often
male, expressed CD15, and had lower expression of CD20.7
In a retrospective multicenter study of 112 patients with GZL studied by Evens and
colleagues,6 the immunophenotype of GZL seemed concordant with prior studies. In
addition, there was no apparent significant difference in immunophenotype between
MGZL and NMGZL. The most prevalent immunophenotypes were CD201 (93%),
CD301 (89%), CD79a1 (78%), Pax51 (98%), Oct21 (96%), and MUM11 (100%).

Table 1
Morphologic and immunophenotypic features of GZL and related pathologic spectrum

Feature PMLBCL DLBCL cHL GZL

Morphology
RS cells and variants Rare Rare Yes Yes in areas
HL background No No Yes Yes in areas
Sheets of large cells Yes Yes Syncytial variant Yes in areas
Fibrous bands No No Yes Uncommon
“Alveolar” fibrosis Yes No No Sometimes
Necrosis Yes Yes Common Common
Neutrophilic infiltrates No No Yes No
Immunophenotype
CD20 1 1 /1 /1 or 1
CD30 1 weak, variable /1 1 1 or 1/
CD15 /1 1/ 1/
CD79a 1 1 /1 /1 or 1
PAX5 1 1 1 (weak) /1 or 1
Bcl6 1/ 1/ 1/ (weak) 1/ variable
OCT2 1 1 /1 1/ or 1
MUM1/IRF4 1/ 1 ABC type 1 1
CD45 1 1 1/ or 1
EBV (EBER) /1 /1 /1

Abbreviations: 1, all cases positive; 1/ , most cases positive; /1, some cases positive; , all cases
negative; ABC, activated B-cell type; cHL, classical Hodgkin lymphoma; DLBCL, diffuse large B-cell
lymphoma; EBER, Epstein-Barr virus–encoded small mRNAs (by in situ hybridization); EBV, Epstein-
Barr virus; HL, Hodgkin lymphoma; RS, Reed-Sternberg.
 Gray Zone Lymphoma 1255

CD15 (46%) and CD45 (67%) staining were more variable, whereas only 11% and
28% of patients had CD10 and Epstein-Barr virus positivity, respectively (the latter
by EBER in situ hybridization staining). It should be highlighted, however, that central
pathologic review was not performed for this retrospective multicenter study; patho-
logic work-up and diagnoses were completed at the local academic centers.
Altogether, diagnosis of GZL remains challenging; work-up and/or consultation at a
center with hematopathology expertise of this particular entity is highly recommen-
ded. We advocate detailed review of B-cell markers (eg, CD20, PAX5, CD79a,
BCL6, BOB1, OCT2) along with CD30, CD15, and MUM1 when GZL is suspected.
In situ hybridization for Epstein-Barr virus should also be included. Collectively, the
diagnosis of GZL should be entertained in cases with transitional morphology and
discordant immunophenotype if strong expression of CD20 is identified in cHL or
strong CD15 expression is seen in an otherwise typical DLBCL.

CLINICAL PRESENTATION

The original reports describing GZL contained patients primarily with mediastinal dis-
ease presenting clinically similarly to PMBCL (Table 2).1,10–12 Subsequent reports of
GZL have shown a distinctive nonmediastinal systemic presentation.6,9 Wilson and
colleagues13 recently reported outcomes among 24 patients with MGZL diagnosed
and treated over an approximate 20-year period at the NIH. The median age of these
patients was 33 years and approximately two-thirds were male. Of these patients,
nearly half had a bulky mediastinal mass greater than 10 cm, whereas a minority of
patients had extranodal involvement and only 13% had stage IV disease.
In terms of comparison of MGZL with NMGZL, patients with NMGZL presented
significantly less often with bulk disease, but they were significantly older and pre-
sented more often with extranodal involvement and advanced-stage disease in the se-
ries by Evens and colleagues.6 In part because of these latter features, patients with
NMGZL had significantly worse prognostic scores (International Prognostic Scoring
system and International Prognostic Index [IPI]) compared with patients with MGZL.

PRIMARY MANAGEMENT

There are no standard management guidelines for GZL. In part this is because of the
small number of analyses examining therapeutic approaches for GZL and challenges
in diagnosis. From the aforementioned prospective NIH study, 24 patients with MGZL
were treated with dose-adjusted etoposide, prednisone, oncovin, cyclophosphamide,
doxorubicin, and rituximab (DA-EPOCH-R) and outcomes were compared with simi-
larly treated patients with PMBCL.13 Patients with MGZL had significantly inferior out-
comes compared with PMBCL (5-year event-free survival 62% vs 93%, and 5-year
overall survival [OS] 74% vs 97%, respectively).13,14
Among the 112 patients with GZL reported by Evens and colleagues treated across
19 North American centers, the two most common treatment approaches were
CHOP-R and doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD)  rituxi-
mab with a handful of patients treated with DA-EPOCH-R. Approximately two-thirds
of patients received rituximab as part of frontline therapy. The overall response rates
and complete remission rates for all patients were 71% and 59%, respectively, with
33% of patients having primary refractory disease (no differences between MGZL
and NMGZL). The overall response rates and complete remission rates for patients
who received rituximab as part of front-line therapy were 82% and 73%, respectively,
versus 59% and 43%, respectively, without rituximab (P 5 .02 and P 5 .008,
respectively).
 1256
Pilichowska et al
Table 2
Clinical characteristics from GZL series

Mediastinal Advanced
Author, Year No. Pts Median Age, y (Range) M:F Disease, % B Symptoms, % Stage, % Bulky Disease, % PFS/EFS, %
Traverse-Glehen et al,1 2005a 21 31 (13–62) 0.90 100 NR NR NR NR
Oschlies et al,10 2011 4 14 (11–18) 3.0 100 50 100c NR 0
Eberle et al,9 2011a 18 29 (16–51) 2.0 100 NR NR NR NR
Eberle et al,9 2011a 9 55 (26–91) 0.50 0 NR NR NR NR
Gualco et al,11 2012 10 34 (15–85) 1.22 80 NR 20 NR 50
Wilson et al,13 2014a 24 33 (14–59) 1.67 100 NR 13 46 62
Evens et al,6 2015b 48 37 (60–93) 1.4 100 45 13 44 46
Evens et al,6 2015b 64 51 (60–84) 2.37 0 63 81 8 36

Minimum four patients.

Abbreviations: EFS, event-free survival; NR, not reported; OS, overall survival; PFS, progression-free survival.
a
There is some overlap of patients studied in these series.
b
Mediastinal and nonmediastinal GZL cases from the same series are separated to highlight differences among these clinical entities.
c
St. Jude’s stage of disease.
 Gray Zone Lymphoma 1257

At 31-month median follow-up, the 2-year progression-free survival (PFS) and OS

for all patients were 40% and 88%, respectively (Fig. 2). Two-year PFS rates for pa-
tients with stage I/II versus III/IV disease were 54% versus 30%, respectively, and
the corresponding OS rates were 94% versus 76%, respectively. Interestingly, despite
a comparative preponderance of early stage disease and lower IPI, the survival rates
were not statistically different for MGZL versus NMGZL patients (see Fig. 2). In terms
of frontline chemotherapy regimens in this retrospective series, PFS rate seemed
significantly inferior for patients treated with ABVD  rituximab versus treatment
with a DLBCL-based regimen (ie, CHOP  R and DA-EPOCH-R) with corresponding
2-year rates of 23% versus 52%, respectively (see Fig. 2). This finding persisted on
multivariable Cox regression, including controlling for IPI and use of rituximab.

ROLE OF RADIOTHERAPY

There are minimal prospective and no randomized data regarding the benefit of radio-
therapy in GZL. In the MGZL DA-EPOCH-R series by Wilson and colleagues,13 use of
radiotherapy is not standard practice with this therapeutic regimen for PMBCL pa-
tients.15 Treatment paradigms for GZL in the study by Evens and colleagues6 more
often used adjunctive consolidative radiotherapy with early stage and/or bulky dis-
ease; improved outcomes were not seen in patients who received radiotherapy. How-
ever, caution should be applied to this finding given the retrospective nature and
relatively small patient numbers. Notably, relapses localized in the mediastinum
treated with radiotherapy may be curative. Wilson and colleagues13 demonstrated
that 44% (four of nine) of patients achieved continuous remission with salvage radia-
tion therapy alone. It is not known if outcomes would be improved for MGZL if adjuvant
radiotherapy were more routinely used in conjunction following DA-EPOCH
chemotherapy.

PROGNOSTICATION

The prospective DA-EPOCH-R study by Wilson and colleagues13 identified PMBL-like

tumor morphology, expression of CD15 on malignant cells, and presence of CD681
tumor-infiltrating dendritic cells as makers for poor prognosis in MGZL. The gene
signature encoding for dendritic cell–specific intracellular adhesion molecule-3-
grabbing nonintegrin (DC-SIGN) is a marker of dendritic cells and activated macro-
phages16; in this series, cases with greater than or equal to one positive tumor
associated-dendritic/activated macrophage cells by immunohistochemistry was
associated with markedly inferior outcomes (ie, OS of 52% vs 100%; P 5 .0025). Simi-
larly to cHL low peripheral blood absolute lymphocyte count in MGZL was associated
with inferior event-free survival and OS, whereas tumor mass size and expression of
CD20 were not prognostic.
Prognostic factors associated with inferior PFS on univariable analyses in the Evens
and colleagues6 series included poor performance status, increased lactate dehydro-
genase, anemia, and advanced stage disease, whereas only stage was prognostic for
OS. Both the IPI and International Prognostic Scoring system were prognostic for PFS
and OS (as categorical and continuous variables). On multivariable analyses, perfor-
mance status was the most dominant prognostic factor for PFS, whereas increased
lactate dehydrogenase was borderline. For OS, advanced stage disease was the
only significant prognostic factor (hazard ratio, 4.89; P<.05). Interestingly, CD20 pos-
itivity in malignant cells also independently predicted PFS on Cox regression control-
ling for rituximab treatment (ie, 88% of CD20-negative patients with GZL experienced
progressive disease).
1258

Fig. 2. GZL outcomes. The 2-year (A) PFS and OS rates for patients with mediastinal versus
nonmediastinal (systemic) GZL from a recently reported multicenter retrospective series.6
The outcomes (B) based on frontline therapeutic regimen; 2-year PFS for patients who
received a standard frontline DLBCL regimen (ie, CHOP  R and DA-EPOCH-R) versus
ABVD  R were 52% versus 23%, P 5 .03. Kaplan-Meier curves (C) for patients who received
rituximab as a part of frontline therapy versus not; 2-year PFS were 51% versus 22%, respec-
tively, P 5 .01. (Adapted from Evens AM, Kanakry JA, Sehn LH, et al. Gray zone lymphoma
with features intermediate between classical Hodgkin lymphoma and diffuse large B-cell
lymphoma: characteristics, outcomes, and prognostication among a large multicenter
cohort. Am J Hematol 2015;90(9):781; with permission.)
 Gray Zone Lymphoma 1259

RELAPSED/REFRACTORY DISEASE

Among available data, the ability to salvage patients with GZL with relapsed/refractory
disease seems good. In the series by Evens and colleagues,6 the median time to
relapse for patients with GZL was 7 months (range, 1–64 months); most patients
were treated with standard combination chemotherapy salvage regimens. Further-
more, most patients underwent hematopoietic stem cell transplantation (HSCT), usu-
ally autologous. The 2-year OS rate for patients with relapsed/refractory GZL who
underwent HSCT was 88% versus 67% for patients who did not. This finding persisted
on multivariable analyses controlling for IPI and response to pre-HSCT salvage ther-
apy.6 This improvement likely reflected in part that fit patients with chemotherapy-
sensitive disease underwent SCT. Nevertheless, given these findings, there should
be consideration for salvage combination chemotherapy followed by autologous
HSCT for patients with relapsed/refractory GZL. Additionally, as described before
for select cases with localized relapse in the mediastinum, there may be consideration
for radiotherapy ( chemotherapy) without HSCT.13

SUMMARY

Since its original description in 20051 and its inclusion in the World Health Organization
in 2008,2 there has been important biologic and clinical knowledge gained about GZL.
Despite the increased recognition of this disease as a pathologic and clinical entity, it
is important to highlight that diagnosis is difficult and requires involvement of hemato-
pathology with disease expertise in this area. Furthermore, there remains a critical
need for pathologic harmonization and consensus diagnostic criteria.
It is also fully recognized that patients with GZL may present with primary medias-
tinal localization or with systemic disease. Pathology or biology differences between
these varied clinical presentations remains to be defined. Regardless of clinical pre-
sentation, patients with GZL have high relapse rates, especially compared with pa-
tients with PMLCL and cHL. Patients with GZL seem to be salvaged fairly
successfully with salvage therapy, especially when including HSCT. Further examina-
tion toward defining the most optimal chemotherapy regimen and the appropriate
timing of HSCT is essential.
Off of a clinical trial, we recommend DLBCL-directed therapy (eg, R-CHOP or DA-
EPOCH-R) for frontline GZL therapy with consideration for use of adjunctive radio-
therapy for patients with localized and/or bulky disease. Finally, there should be
continued exploration toward the use and integration of novel targeted therapeutic
agents (eg, brentuximab vedotin, new-generation CD20 antibodies, PD-1 inhibitors)
into the treatment paradigm of GZL.

REFERENCES

1. Traverse-Glehen A, Pittaluga S, Gaulard P, et al. Mediastinal gray zone lym-

phoma: the missing link between classic Hodgkin’s lymphoma and mediastinal
large B-cell lymphoma. Am J Surg Pathol 2005;29(11):1411–21.
2. Swerdlow SH, Campo E, Harris NL, et al. WHO classification of tumours of hae-
matopoietic and lymphoid tissues. 4th edition. Lyon (France): International
Agency for Research on Cancer; 2008.
3. Grant C, Dunleavy K, Eberle FC, et al. Primary mediastinal large B-cell lym-
phoma, classic Hodgkin lymphoma presenting in the mediastinum, and medias-
tinal gray zone lymphoma: what is the oncologist to do? Curr Hematol Malig Rep
2011;6(3):157–63.
1260 Pilichowska et al

4. Garcia J, Mollejo M, Fraga M, et al. Large B-cell lymphoma with Hodgkin’s fea-
tures. Histopathology 2005;47(1):101–10.
5. Eberle FC, Salaverria I, Steidl C, et al. Gray zone lymphoma: chromosomal aber-
rations with immunophenotypic and clinical correlations. Mod Pathol 2011;24(12):
1586–97.
6. Evens AM, Kanakry JA, Sehn LH, et al. Gray zone lymphoma with features inter-
mediate between classical Hodgkin lymphoma and diffuse large B-cell lym-
phoma: characteristics, outcomes, and prognostication among a large
multicenter cohort. Am J Hematol 2015;90(9):778–83.
7. Dunleavy K, Wilson WH. Primary mediastinal B-cell lymphoma and mediastinal
gray zone lymphoma: do they require a unique therapeutic approach? Blood
2015;125(1):33–9.
8. Eberle FC, Jaffe ES. XII. Gray zone lymphomas: a biological experiment, and a
challenge for diagnosis and management. Ann Oncol 2011;22(Suppl 4):iv64–6.
9. Eberle FC, Rodriguez-Canales J, Wei L, et al. Methylation profiling of mediastinal
gray zone lymphoma reveals a distinctive signature with elements shared by
classical Hodgkin’s lymphoma and primary mediastinal large B-cell lymphoma.
Haematologica 2011;96(4):558–66.
10. Oschlies I, Burkhardt B, Salaverria I, et al. Clinical, pathological and genetic fea-
tures of primary mediastinal large B-cell lymphomas and mediastinal gray zone
lymphomas in children. Haematologica 2011;96(2):262–8.
11. Gualco G, Natkunam Y, Bacchi CE. The spectrum of B-cell lymphoma, unclassi-
fiable, with features intermediate between diffuse large B-cell lymphoma and
classical Hodgkin lymphoma: a description of 10 cases. Mod Pathol 2012;
25(5):661–74.
12. Minami J, Dobashi N, Asai O, et al. Two cases of mediastinal gray zone lym-
phoma. J Clin Exp Hematop 2010;50(2):143–9.
13. Wilson WH, Pittaluga S, Nicolae A, et al. A prospective study of mediastinal gray-
zone lymphoma. Blood 2014;124(10):1563–9.
14. Dunleavy K, Pittaluga S, Shovlin M, et al. Untreated primary mediastinal B-cell
(PMBL) and mediastinal grey zone (MGZL) lymphomas: comparison of biological
features and clinical outcome following DA-EPOCH-R without radiation. Ann On-
col 2011;134(22).
15. Dunleavy K, Pittaluga S, Maeda LS, et al. Dose-adjusted EPOCH-rituximab ther-
apy in primary mediastinal B-cell lymphoma. N Engl J Med 2013;368(15):
1408–16.
16. Geijtenbeek TB, Krooshoop DJ, Bleijs DA, et al. DC-SIGN-ICAM-2 interaction me-
diates dendritic cell trafficking. Nat Immunol 2000;1(4):353–7.