1993 81: 2102-2109   

 D Shibata, LM Weiss, AM Hernandez, BN Nathwani, L Bernstein and AM Levine infected with the human immunodeficiency virus [see comments]Epstein-Barr virus-associated non-Hodgkin's lymphoma in patients

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

reserved.Copyright 2011 by The American Society of Hematology; all rights900, Washington DC 20036.weekly by the American Society of Hematology, 2021 L St, NW, Suite Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published    

For personal use only. by guest on October 9, 2013. bloodjournal.hematologylibrary.orgFrom For personal use only. by guest on October 9, 2013. bloodjournal.hematologylibrary.orgFrom For personal use only. by guest on October 9, 2013. bloodjournal.hematologylibrary.orgFrom For personal use only. by guest on October 9, 2013. bloodjournal.hematologylibrary.orgFrom For personal use only. by guest on October 9, 2013. bloodjournal.hematologylibrary.orgFrom For personal use only. by guest on October 9, 2013. bloodjournal.hematologylibrary.orgFrom For personal use only. by guest on October 9, 2013. bloodjournal.hematologylibrary.orgFrom For personal use only. by guest on October 9, 2013. bloodjournal.hematologylibrary.orgFrom For personal use only. by guest on October 9, 2013. bloodjournal.hematologylibrary.orgFrom

Epstein-Barr Virus-Associated Non-Hodgkin’s Lymphoma in Patients Infected With the Human Immunodeficiency Virus

By Darryl Shibata, Lawrence M. Weiss, Antonio M. Hernandez, Bharat N . Nathwani, Leslie Bernstein, and Alexandra M. Levine

Lymphoproliferations associated with Epstein-Barr virus (EBV) commonly arise in settings of immune dysfunction, including human immunodeficiency virus (HIV) infection. In this study, EBV was associated with 39 of 59 (66%) HIV- related systemic lymphomas. Unlike the lymphoprolifera- tions that arise in the setting of transplantation, the HIV- related lymphomas were monoclonal, as evaluated by lg heavy chain rearrangements and EBV termini analysis, and associated (40%) with c-MYC rearrangements. Further- more, analysis of multiple lymphoma tissues from one au- topsy showed evidence that a single lymphoma clone was responsible for dissemination. The latent EBV nuclear an- tigen (EBNA-1) transcripts detected in the HIV-related lymphomas were characteristic of the pattern found in

NFECTION WITH the human immunodeficiency virus I (HIV) is associated with a greatly increased risk of de- veloping non-Hodgkin’s lymphoma (NHL). This risk is cur- rently estimated at approximately 60- to 100-fold, and in- creases with longer survival and decreasing CD4 counts.’,2 Both reactive and neoplastic lymphoproliferations have been described in patients with HIV infection, which may be anal- ogous to the various B-cell lymphoproliferative disorders that arise in other states of immune dysregulation, such as organ transplantation with iatrogenic immunosuppression. How- ever, the precise molecular characteristics of HIV-related lymphomas appear to differ from those described in the transplantation setting. Thus, whereas most transplant-related lymphoproliferations contain Epstein-Barr viral (EBV) se- q u e n c e ~ , ~ . ~ studies of HIV-infected patients with systemic lymphoma have demonstrated a significantly lower frequency of EBV

In a prior study of lymph node biopsies from enlarged nodes of HIV-infected patients with reactive persistent gen- eralized lymphadenopathy, the detection of EBV DNA in the reactive biopsy was associated with a higher risk of de- veloping lymphoma in time, or the presence of concomitant lymphoma at another site.’’ A high percentage of the lym- phomas that developed were also EBV associated, although the number of cases was small. Therefore, to better charac- terize these findings, we studied a larger group of patients

From the Departments ofPathology, Preventive Medicine, and He- matology, Los Angeles County-University of Southern Calijbrnia Medical Center, Los Angeles; and the Department of Pathology, City of Hope National Medical Center, Duarte, CA.

Submitted September 16. 1992; accepted November 24, 1992. Supported by Grants No. CA50341 and CAS0850 from the National

Cancer Institute. Address reprint requests to Darryl Shibata, MD, LAC-USC Medical

Center, 1200 N State St, No. 736, Los Angeles, CA 90033. The publication costs ofthis article were defrayed in part b-v page

charge payment. This article must therefore be hereby marked “advertisement” in accordance with I8 U.S.C. section I734 solely to indicate this fact.

0 I993 by The American Society if Hematology. O006-4971/93/8108-0012$3.00/0

Burkitt lymphoma (91 EBNA1) and not in transplant-related lymphoproliferations. However, unlike Burkitt lymphoma, EBV latent membrane-associated protein (LMP) transcripts were also detected, thereby constituting an EBV expression pattern (91 EBNAl +, LMP+) not previously observed in B- cell lymphomas. These findings demonstrate a high fre- quency of EBV-associated lymphomas in the setting of HIV infection that are distinct from the lymphoproliferations that arise during iatrogenic transplant-associated immuno- suppression or in the general population. However, it is also apparent that HIV-related lymphomas are biologically het- erogeneous, which may reflect the multiple mechanisms or steps necessary for eventual malignant transformation. 0 1993 by The American Society of Hematology.

with HIV-related lymphoma to determine the presence and expression of EBV sequences within lymphoma tissue.

MATERIALS AND METHODS

Subjects and specimens. Tissues were derived from subjects di- agnosed between 1984 and I992 with systemic intermediate- or high- grade lymphoma, living in the County of Los Angeles, as part of a larger epidemiologic study of non-Hodgkin’s lymphoma. Fifty-nine cases were diagnosed with acquired immunodeficiency syndrome (AIDS)-related lymphoma, after demonstration of seropositivity by HIV enzyme-linked immunosorbent assay (ELISA) and confirmatory Western blot. Thirty-seven cases occurred in subjects who were found to be HIV seronegative; tissues from these patients were used as “controls” for the HIV-related lymphoma cases. No case of primary central nervous system (CNS) lymphoma was studied. Formalin- or B5-fixed, paraffin-embedded tissue was available on all cases, con- stituting criteria for entry into this investigation. In addition, frozen tissues were also available for further study on I5 cases of HIV-related lymphoma, and 2 I cases on non-HIV-related lymphoma. The lym- phomas were classified using the Working Formulation” by three of the authors (D.S., A.M.H., and B.N.N.). Clinical data were obtained by review of the patients’ medical records. Fisher’s exact test was used for statistical analysis.

DNA was extracted from the formalin-fixed, paraffin-embedded tissue sections or fresh-frozen tissue (see below) from each lymphoma specimen, and was examined for EBV sequences as previously described.’’ A PCR assay against the EBV nuclear antigen- 1 (EBNA- I ) gene was used to screen all spec- imens. Both EBV-positive and -negative specimens were all positive for the human @-globin sequence, verifying the suitability of the ex- tracted DNA for PCR analysis. The EBV-positive specimens were further amplified with primers directed to a polymorphic region of EBNA-2, which differs between types A and B EBV variants.” The amount of EBV in each positive specimen was also estimated by serial 10-fold dilutions followed by PCR for the @-globin and the EBV EBNA-1 sequence.

ISH was performed on the EBV PCR- positive specimens using a biotin-labeled EBV-encoded RNA (EBER- 1 ) oligomer probe as previously described.I4 In addition, several EBV PCR-negative lymphomas were subjected to ISH and were always negative.

Tissue frozen at -70°C was available in selected cases. High molecular weight DNA was extracted using a nonorganic extraction kit (Oncor, Gaithersburg, MD). The DNA was digested with restriction enzymes (Bethesda Research Labora-

Polymerase chain reaction (PCR).

In situ hybridization (ISH).

Southern blot anal.vsis.

2102 Blood, Vol81, No 8 (April 15). 1993: pp 2102-2109

EBV-ASSOCIATED HIV-RELATED LYMPHOMA 2103

tones [BRL], Gaithersburg, MD) overnight at 37”C, separated on a 0.7% agarose gel, and then Southern blotted using standard capillary techniquesL5 to a nylon filter (Magnagraph; MSI, Westboro, MA). The DNA was cross-linked to the filter using UV irradiation and then hybridized overnight at 42°C to various DNA probes. The filters were washed with a final stringency of O S X SSPE/O. 1% sodium do- decyl sulfate (SDS) at 55°C for l hour. Autoradiography was per- formed for periods ranging from overnight to 2 weeks.

Ig heavy chain rearrangements were detected using Hind111 and BamHI/HindlII digestions with a 5.4-kb HindIIl-BamHI JH probe (Oncogene Science, Uniondale, NY). T-P chain receptor rearrange- ments were analyzed using BamHI or EcoRl digestions with a 0.4- kb constant region probe (Oncogene Science). c-MYC rearrangements were detected using EcoRI and Hind111 digestions with a third exon 1.4-kb Cla I-EcoRI probe (Oncor). EBV termini structure was ana- lyzed using BamHI digestion and a 1.9-kb Xho I fragment located near the right EBV terminus (generously provided by Dr N. Raab- Traubf6). The DNA probes were labeled with ”P-dATP using the random hexamer method.15

Detection ofEBVmRNA. Reverse PCR was performed to detect the group 1 (gl) or group 3 (g3) EBNA-1 RNA as previously de- scribed.” A single frozen section was placed in a sterile 1.5 mL mi- crofuge tube. Poly A+ RNA was isolated using oligo(dT)-cellulose columns with a commercial kit (Quickprep; Pharmacia, Piscataway, NJ). First-strand cDNA synthesis was performed on approximately 15% of the RNA from each section using Moloney murine leukemia virus reverse transcriptase (BRL) in standard PCR buffer at 37°C for 30 minutes.’* The “K’ primer” was used to prime the first-strand synthesis. The cDNA was split into two different aliquots and then subjected to PCR with either the “Q’ and “K’ primers to detect the g l or the “Y3” and “ K primers to detect 83 EBNA-1 mRNA. A total of 50 PCR cycles was performed, with denaturation at 95°C for 60 seconds, annealing at 54°C for 50 seconds, and extension at 72°C for 60 seconds, with a final extension at 72°C for 7 minutes. The PCR products were separated on a 2.5% Nuseive (FMC, Rock- land, ME), 0.5% agarose gel and Southern blotted to a nylon filter (Genetrans 45; Plasco, Woburn, MA). The filter was hybridized to a 32P end-labeled “U” exon oligomer probet7 and washed in 2X SSPE at room temperature. Autoradiography was performed for 3 to 72 hours. A positive control of poly A+ RNA isolated from the EBV- positive Raji cell line demonstrated high g3 EBNA-I mRNA levels and very low gl EBNA-I mRNA levels.

The EBV mRNAs for latent membrane-associated protein 1 (LMPI), LMP2A, and LMP2B were detected by reverse PCR as above, using the same aliquots of poly A+ RNA with primers de- scribed by Brooks et aI.l9 The primers for LMPl were CTTCAGAA- GAGACCTTCTCT (upstream) and ATACCTAAGACAAGTA- AGCA (downstream and used for cDNA synthesis). The oligomer ACAATGCCTGTCCGTGCAAA was used as an LMPl probe. LMP2 mRNA was detected using a common downstream primer for cDNA synthesis and then a different upstream primer specific for either LMP2A or LMP2B. An oligomer common to both LMP2A and LMP2B was used as a labeled probe.I9 The PCR was performed as above with an annealing temperature of 50°C and 50 cycles.

A lymphoma was con- sidered EBV associated only if EBV could be detected by at least two methods including PCR, demonstration of labeling of the tumor cells by ISH, or clonal infection using the EBV termini probe.

Criteria for EB V-associated lymphoma.

RESULTS

EBV was detected in specimens from 39 of the 59 (66%) HIV-related lymphoma cases by PCR, along with ISH or Southern blot assays for EBV termini (Tables 1 and 2). In contrast, EBV was detected in tissues from only 2 of the 37

Table 1. EBV Analysis of HIV-Positive and HIV-Negative Lymphoma

EBV Positive EBV

NHL Type Negative Total Type A Type B Unknown

HIV positive

SNC IBL Subtotal

LC SNC IBL Subtotal

Frozen specimens

Fixed specimens

Total

HIV negative

LC SNC IBL Subtotal

SNC IBL Subtotal

Frozen specimens

Fixed specimens

Total

2 2 4

5 10

1 16 20

12 4 4

20

7 8

15 35

6 5

11

6 10 12 28 39

0 1 1 2

0 0 0 2

5 1 0 2 3 0 7 4 0

4 2 0 6 2 2 8 2 2

18 6 4 25 10 4

0 0 0 1 0 0 0 1 0 1 1 0

0 0 0 0 0 0 0 0 0 1 1 0

Abbreviations: I8L. immunoblastic lymphoma; SNC. small noncleaved lymphoma; LC, large cell lymphoma.

( 5 % ) non-HIV-related lymphoma cases (P < .00001). A specimen was considered EBV associated only if EBV was detected by at least two methods; thus, four HIV-related cases in which EBV was detected by PCR but could not be localized to the lymphoma cells by ISH were considered EBV negative. Large numbers of EBV sequences were estimated to be pres- ent in the EBV-positive cases, based on serial dilutions fol- lowed by PCR (not shown). In the EBV-positive specimens, EBER-1 R N A was localized to the nucleus as previously reported2’ (Fig 1). The large number of EBER-I R N A mol- ecules expressed per infected cell (up to 10’) facilitated de- tection by ISH.

The EBV-positive, HIV-related cases did not have dis- tinctive histologic features and included 6 of I 1 (55%) cases of diffuse large cell, 16 of 28 (57%) cases of small noncleaved, and 17 of 20 (85%) cases of immunoblastic lymphoma (Table 3). None of the HIV-related small noncleaved lymphomas had the classic histology of Burkitt lymphoma (BL). In this study, which excluded primary CNS lymphoma, the majority of the HIV-related systemic lymphomas were extranodal (59%). Approximately equal proportions of the lymphomas arising in nodal sites (67%) and extranodal sites (66%) were EBV associated.

EBV may be subclassified into two groups, type A and type B, based on sequence polymorphisms in the EBNA-2 gene.I3s2l The EBV type in the 39 positive HIV-related cases was determined to be type A in 25 (64%) cases and type B in 10 (26%) cases. N o mixed infections were detected. The EBV type could not be determined in four cases. The two

2104 SHIBATA ET AL

Table 2. Analysis of Frozen Tissue From Diffuse Large Cell Lymphomas

EBV RNA

Case Type J H TB MYC EBV PCR EBV ISH EBV Type EBV Termini EBNA 1 LMP

HIV positive 1 2 3 4 5 6 7 8 9

10 11 12 13 14 15

HIV negative A B C D E F G H I J K L M N 0 P 0 R S T U V

SNC SNC SNC IBL IBL IBL SNC SNC IBL IBL SNC IBL SNC IBL SNC

SNC IBL SNC IBL IBL IBL LC IBL SNC SNC SNC LC LC LC LC LC LC LC LC LC LC LC

R R R R R R R R R R R R R R R

R R R R R R R R R R R R R R R R R R R R R R

G G G G G

G G

G

G

G G G

G G G G G G G G G

G

R G R R G R R R G G G G G G G

R G R R G G G G G G G G G G G G G G G G G G

+ +

A A A B A B A B B A A

A B

MCL 9’ + MCL g l + MCL 91 + MCL g l + MCL NA NA MCL MCL MCL MCL MCL MCL - - -

-

Abbreviations: SNC. small noncleaved lymphoma; IBL, immunoblastic lymphoma; LC. large cell lymphoma; R, rearranged; G, germline configuration; NA, study not adequate; +, positive; -, negative; g l , group 1 EBNAl mRNA; MCL, monoclonal episomal EBV terminal configuration.

EBV-associated lymphomas from the non-HIV-infected pa- tients contained type A in one case and type B in the other.

The g 1 pattern of EBNA- 1 mRNA expression was detected in four EBV-associated, HIV-related lymphomas (Fig 2). There was no evidence of the g3 expression pattern. In con- trast, mRNA isolated from the Raji cell line demonstrated a predominance of the g3 EBNA-1 message with Scant gl expression. Using reverse PCR, the mRNAs for LMPl or LMP2A were also detected. No LMP2B transcripts were de- tected. Attempts to demonstrate LMP by immunohisto- chemistry were technically unsatisfactory (not shown).

The clonality of EBV infection and structure of the EBV genome was then analyzed by determining the number and size of fused terminal fragments.16 This analysis was per- formed on 12 EBV-positive cases from which sufficient frozen

tissue was available, using an EBV-terminal probe and a BumHI digestion (Fig 3). A single band greater than 8 kb, indicating one species of EBV episome, was detected in each case. There was no evidence of linear EBV genomes except for very faint smears between the 4- and 6-kb regions. These results are consistent with a monoclonal expansion of the EBV-infected lymphoma cells. Further, active viral replica- tion, if present at all, comprised only the minority of the EBV genomes.

The 15 HIV-related and 22 non-HIV-related lymphoma specimens with frozen tissue available were further studied by Southern blot analysis. In all of these specimens, clonal Ig heavy chain rearrangements were detected. Rearrangement of the T-/3 receptor was not detected in the 22 cases examined. Using a third exon probe, c-MYC rearrangements were de-

EBV-ASSOCIATED HIV-RELATED LYMPHOMA 2105

Fig 1. Positive ISH with the EBV EBER-1 probe demonstrating the expected nuclear localization (case 6, original magnification X 600).

tected in 6 (40%) HIV-related lymphomas. The HIV lym- phomas with rearranged c-MYC genes were all EBV positive. In contrast. c-MYC rearrangements were detected from only 3 (14%) of the non-HlV-related lymphomas and only I of these was EBV positive.

In one autopsy case (case 1 I). lymphoma tissues from multiple sites (lymph nodes, rectal mass, and vesical mass) were available for Ig heavy chain and EBV termini clonal analysis. In every involved site, the identical monoclonal Ig heavy chain and EBV termini rearrangements were detected (Fig 4). An additional. very faint higher molecular weight EBV-terminal fragment was also detected at one lymphoma tissue site (hilar node, not visible in Fig 4). These data suggest that dissemination of lymphoma in this patient occurred pri- marily through the clonal expansion of a single transformed EBV-infected B-lymphoid cell.

DISCUSSION

The majority (66%) of HIV-related systemic lymphomas in this study were EBV associated. This frequency is some- what higher than that noted in the majority of prior studies, in which 38% to 50% of cases were found to be EBV asso- ciated.6.8-'" Higher frequencies of EBV involvement, similar to our own, have also been noted.'.*' The explanation for the differences between these studies is unclear, but may involve differences in patient population. histologic types of lym-

Table 3. Summary of EBV Association in HIV-Positive and HIV-Negative Lymphoma Cases

HIV Negative HIV Positive

Type EBV- EBV+ %Positive EBV- EBV+ %Positive

LC 12 0 0 5 6 55 SNC 11 1 8 12 16 57 IBL 12 1 8 3 17 85 Total 35 2 5 20 39 66

Abbreviations: LC, large cell lymphoma; SNC, small noncleaved lym- phoma; IBL, immunoblastic lymphoma.

phoma. or anatomical sites of disease. For example. the vast majority of HIV-related primary CNS lymphomas are EBV as~ociated.~-'.*~ However, no case of primary CNS lymphoma was examined in the current study and we found no difference in the frequency of EBV-positive lymphomas arising from extranodal or nodal sites. The frequency (59%) of systemic extranodal lymphoma observed in this study is similar to that found in other reports.'

Similar to the lack of uniformity in the setting of HIV- related lymphoma, different frequencies of EBV association have also been found in non-HIV-related lymphoma occur- ring throughout the world. Thus, EBV is present in nearly all examples of BL from Africa, but is much less frequent (10% to 20%) in cases from the United Stateszs Studies of similar lymphomas from other geographic regions such as South America have demonstrated an intermediate (50%) frequency of EBV association.'6 The different frequencies of EBV involvement observed in BL from diverse geographic areas may be a reflection of diverse mechanisms of lym- phomagenesis, which may or may not require preceding EBV infection. The relatively low frequency of c-MYC rearrange- ments (40%) detected in this study of HIV-related lympho- proliferations, compared with prior studies? may also reflect this diversity. More detailed analysis of the specific c-MYC and Ig heavy chain in these lymphomas may allow better comparisons between groups. In our population from Los Angeles, it is clear that lymphomas that occur in the setting of underlying HIV infection are more likely to be associated with EBV than similar lymphomas that occur in the absence of HIV infection.

The HIV-related lymphomas in this study were all diffuse lymphomas. Similar to other studies,'.'." the majority were high-grade small noncleaved or immunoblastic lymphomas. Of note, the presence of EBV was not restricted to a specific histologic type of lymphoma. Thus, EBV sequences were present in more than half of the diffuse large cell, small non- cleaved, and immunoblastic lymphomas. As observed by others. the majority (85%) of the immunoblastic lymphoma cases in the current study were EBV associated."'.'" However,

2106 SHIBATA ET AL

P R I M E R S QIK !!J P R I M E R S Y3/K $ (gp3 mRNA) + - A (gpl mRNA)

A B C D E F S C A B C D E F S ‘

Y @

5 2 $ 2 F-= - -4 -268

- 4-239 BP

B

4-231 “ I 4-153

LMPl

@ 0 4-260

LMPPA

8 . 4 4 2 BP

LMPPB Fig 2. (A) PCR products representing the group 1 (gpl) EBNA-

1 mRNA (239 bp) were detected from four EBV-associated, HIV- related lymphomas (A, case 1 ; B, case 2; C, case 3; F, case 4). The group 3 (gp3) EBNA-1 mRNA (268 bp) was detected from the Raji cell line control but not from the specimens. The negative control is poly A RNA extracted from an uninfected T-cell line (CEM). Spec- imen D is an EBV-negative, HIV-related lymphoma (case 12). No EBNA-1 signal was detected from specimen E (EBV-associated. HIV-related lymphoma, case 5). but its RNA appeared degraded (not shown). (B) PCR products from the same specimens demon- strating positive LMPl signals (1 53 bp) in B, C, and F, and positive LMP2A signals (260 bp) in A, B, C, and F. No LMP2B signals (342 bp) were detected. The 231 -bp signal in the LMPl assay is derived from EBV DNA isolated with the RNA because it persists if the reverse transcription step is omitted (with loss of the 153-bp RNA signal).

it is clear that the presence or absence of EBV genomes is not confined to one morphologic type of lymphoproliferation in the setting of HIV infection.

The high frequency of EBV association in the setting of HIV-related lymphoma is not surprising, because EBV ge- nomes have been found within lymphoma tissue in diverse types of underlying immune dysregulation. In the setting of transplantation, almost 100% of the diffuse B-cell lympho- proliferations are EBV elated.^-^ This association may be related to the ability of EBV to infect and immortalize B cells in vitro, yielding lymphoblastoid cell lines.” Of interest, transplant-related lymphoproliferations may have relatively limited malignant potential, because they often regress after reduction or withdrawal of iatrogenic immuno~uppression.~~~~ In contrast to the lymphomas arising in the general popu- lation, the transplant-associated lymphoproliferative disorders may be oligoclonal, multiclonal, or p o l y c l ~ n a l , ~ ~ ~ ~ ~ ~ ’ and chromosomal translocations or c-MYC rearrangements are characteristically absent.32

Of interest. the EBV-related lymphomas in HIV-infected patients appear to be biologically distinct from the transplant- related lymphoproliferations (Table 4). Similar to other stud-

the HIV-related lymphomas in the current study were predominantly monoclonal proliferations, as analyzed by Ig heavy chain rearrangements and EBV termini. The detection of only a single episomal form of EBV DNA suggests that EBV infection preceded transformation. In addition, repli- cative EBV genomes, observed in transplantation-associated lymphopr~liferations?~ were not significantly detected in the current HIV-related lymphomas. Furthermore, unlike the multiple, distinct, EBV-associated clones that arise simulta- neously at different anatomic sites in the setting of trans- plantati~n,’~.~’ the one autopsy case analyzed herein dem- onstrated an identical clone at all involved sites. Thus. although multiple small oligoclonal expansions have been observed in the reactive biopsies and lymphoma tissues from some HIV-infected patients3’ it appears that a single trans- formed clone can be responsible for dissemination of lym- phomatous disease. The additional minor oligoclonal Ig heavy chain rearrangements noted by Pelicci et a135 were not de- tected in our lymphoma biopsies, although our assay may be less sensitive because similar minor clonal rearrangements are typically not detected in the benign hyperplastic lymph node biopsies from our population of HIV-infected patients (data not shown). These findings, along with the detection of c-MYC rearrangements in 40% of our studied cases. in- dicate pathophysiologic mechanisms in HIV-related lym- phomas that are distinct from those found in the setting of iatrogenic immunosuppression and organ transplantation.

A B C D E F G -z -1

W

Fig. 3. Southem blot analysis of EBV termini. The DNA was di- gested with BamHl and hybridized with a 1.9-kb Xho I fragment of the EBV right terminus. The hybridization signals varied widely in this 20-hour exposure, most likely reflecting the variable numbers of EBV genomes per lymphoma cell and the relative proportion of lymphoma cells in the specimens. Shorter exposure periods dem- onstrated a single band in the strongly positive cases. A single faint band was present after a 14-day exposure in lane D. Distinct bands between 4 and 6 kb representing linear DNA were not detected. (A, case B; B, case 1 ; C, case 2; D, case 9; E, case 6; F, case 7; G, case 3)

EBV-ASSOCIATED HIV-RELATED LYMPHOMA 2107

A 3 0 z 3 a a z

0

- f 0 l- a 0 a

c/) v)

z a

2 Y a

v) v)

z a a 0 m W >

3 2 Y a

a a

a 2 W

1

a 8 I

z

4-1 1 KB

4-1 2

4-8

4 4 K B

Fig 4. Southern blot analysis of lymphoma tissue from multiple autopsy sites from case 11. (A) lg heavy chain analysis (Hindlll digestion) demonstrates a single and similar molecular weight rear- ranged band at all lymphoma sites except for the uninvolved liver tissue, which only demonstrates the germline 11 -kb band. (B) EBV termini analysis (&"HI digestion) demonstrates a single and similar molecular weight band at all sites except the uninvolved liver. There was no evidence of smaller linear EBV fragments. A faint, slightly higher molecular weight band was evident in the lymphoma tissue from the hilar lymph node with a longer exposure. The controls demonstrated the expected monoclonal episomal pattern (MONO, single band present on shorter exposure) and a polyclonal pattern (POLY) with linear viral fragments.

EBV-related lymphoproliferative disorders can also be compared in terms of their patterns of latent viral expression. Thus, multiple EBV nuclear and latent membrane-associated antigens (EBNAs 1, 2, 3, and LMPs) are expressed in trans- plant-related lymphoproliferations, in lymphoblastoid cell lines, and in the EBV-associated lymphoproliferations that occur in severe combined immunodeficient This EBV expression pattern has been designated "group 3" (g3).'7.40 A second pattern of EBV expression, present in BL, is characterized by the expression of only the EBNA-I nuclear antigen and is designated "group I" (gl). The EBNA-I mRNA transcripts are different in the two patterns because although they all include EBNA-I sequences, the gl EBNA-

I mRNA arises from a different EBV promoter and is dif- ferentially spliced compared with the g3 EBNA-I mRNA. A PCR assay, using a common downstream primer for cDNA synthesis, can detect and compare the relative abundance of these two mRNAs.I7 The predominant expression of EBNA- 1 mRNA in the current HIV-related lymphomas appears re- stricted to the gl pattern, similar to that found in endemic and sporadic BL. By analogy, expression of gl EBNA-I mRNA is characteristically associated with the absence of other g3 nuclear antigens (EBNA-2 and EBNA-3) because these g3 mRNAs all arise from the same promoters used for g3 EBNA- 1 transcription. However, EBV expression may be heterogeneous in patients with HIV-related lymphoma, be- cause EBNA-2 has been detected in the lymphoma cells of some cases with immunohistochemical method^.^^*^'^^*

Although similar to BL. the patterns of EBV expression in the current HIV-related lymphomas were not identical to the gl model. because LMP transcripts were also detected. This pattern of latent EBV expression (gl EBNAI+, LMP+) has been called "latency 11" and can be induced in vitro by stimulating EBV-infected gl B-lymphoblastoid cell lines!3 This phenotype of latent EBV expression has not previously been observed in NHL. because BL does not express LMP.'7.40 Interestingly, this EBV expression pattern has been observed in Hodgkin's disease and nasopharyngeal c a r ~ i n o m a . ' ~ . ~ ~ ~ ' LMP has transforming proper tie^^^.^^ and its observed expression in the HIV-associated lymphomas may potentially contribute to the process of lymphomagenesis. Further, LMP is an immunogenic cytotoxic target:' which may prevent its expression except in the setting of immunodeficiency.

The pattern of EBV expression observed in the HIV-related lymphomas may explain the detection of EBV types A and B within tumor cells. Similar to previous studies, and anal- ogous to endemic BL, both EBV types A and B were present in the HIV-related lymphomas.' In contrast, sporadic (American) BL is characterized by a predominance of type A." EBV types A and B differ at their EBNA-2 sequences, and lymphocyte transformation is more difficult with type B v i r~s . "~ However, this difference observed in vitro may not be pertinent in HIV-related lymphomagenesis because the EBNA-2 gene does not appear to be strongly expressed.

The contribution of EBV to the etiology and pathogenesis of HIV-related lymphoma is currently unknown, but may be similar to that in BL because the pattern of EBV EBNA- I transcription, timing of infection with respect to transfor-

Table 4. Comparison Between Different Types of EBV-Related Lymphoproliferative Disorders

This Study Transplant- HIV-Related Related Endemic EL Sporadic EL

EBV frequency 66% - 100% - 100% - 10-20% EBNAl mRNA g l 93 g l g l LMP expression Yes Yes No No EBV type A or B Nodata AorB A Clonality Monoclonal Oligoclonal Monoclonal Monoclonal MYC alterations 40% Absent -100% -100%

Abbreviations: EL, Burkitt lymphoma; g l , group 1 EBV EBNA expres- sion pattern; 93. group 111 EBV EBNA1 expression pattern.

2108 SHIBATA ET AL

mation, and dissemination of disease appear analogous. However, the EBV-associated, HIV-related lymphomas are distinct from known examples of NHL because they coexpress g l EBNA-1 and LMP. HIV infection is characterized by a gradual but progressive dysregulation of the immune system and with significant, ongoing antigenic s t im~lat ion.~’ .~’ As noted by others,52 this clinical setting is analogous to the rel- ative immunodeficiency and antigenic stimulation present in the setting of chronic malarial infestation in those regions of Africa endemic for BL.53,54 The evolution of lymphoma in the setting of HIV infection most likely involves multiple steps over a relatively short period of time, which may include EBV infection, c-MYC mutations, and other genetic altera- tions such as p53 i n a ~ t i v a t i o n . ~ ~ Of interest, the molecular genetic aberrations observed in this study and others are not uniform, but rather heterogeneous, and most likely reflect diverse pathogenic mechanisms in the setting of chronic ex- posure to multiple antigenic stimuli, characteristic of patients with underlying HIV infection.

ACKNOWLEDGMENTS

We are indebted to Yuan-Yuan Chen, Debra Hawes, and Zhihua Li for their technical assistance; Dr Adrian Ortega for his assistance in obtaining tissues; and Dr John Sninsky of Roche Molecular Systems for supplying the Taq enzyme used in this study.

REFERENCES I . Levine AM: Acquired immunodeficiency syndrome-related

lymphoma. Blood 803, 1992 2. Beral V, Peterman T, Berkelman R, Jaffe H: AIDS-associated

nomHodgkin’s lymphoma. Lancet 337305, 199 I 3. Hanto DW, Gajl-Peczalska KJ, Frizzera G, Authur DC, Balfour

HH Jr, McClain K, Simmons RL, Najarian JS: Epstein-Barr virus (EBV) induced polyclonal and monoclonal B-cell lymphoproliferative diseases occumng after renal transplantation. Ann SUB 198:356, 1983

4. Hanto DW, Birkenbach M, Fizzera G, Gajl-Peczalska KJ, Sim- mons RL, Schubach WH: Confirmation ofthe heterogeneity of post- transplant Epstein-Barr virus-associated B-cell proliferations by im- munoglobulin gene rearrangement analysis. Transplantation 47:458, 1989

5. Cleary ML, Dorfman RF, Sklar J: Failure in immunological control of the virus infection: Post-transplant lymphomas, in Epstein MA, Achong BG (eds): The Epstein-Barr Virus. New York, NY, John Wiley & Sons, 1986, p 164

6. Subar M, Neri A, lnghirami G, Knowles DM, Dalla-Favera R: Frequent c-myc oncogene activation and infrequent presence of Ep- stein-Barr virus genome in AIDS-associated lymphoma. Blood 72: 667, 1988

7. Ernberg 1, Altiok E: The role of Epstein-Barr virus in lymphomas of HIV-carriers. APMIS Suppl 858, 1989

8. Boyle MJ, Sewell WA, Sculley TB, Apolloni A, Turner JJ, Swanson CE, Penny R, Cooper DA: Subtypes of Epstein-Barr virus in human immunodeficiency virus-associated non-Hodgkin’s lym- phoma. Blood 78:3004, 1991

9. Hamilton-Dutoit SJ, Pallesen G, Franzmann MB, Karkov J, Black F, Skinhoj P, Pedersen C: AIDS-related lymphoma. Am J Pa- tho1 138:149, 1991

IO. Pedersen C, Gerstoft J, Lundgren JD, Skinhoj P, Bottzauw J, Geisler C, Hamilton-Dutoit SJ, Thorsen S, Lisse I, Ralfkiaer E, Pal- lesen G: HIV-associated lymphoma: Histopathology and association with Epstein-Barr virus genome related to clinical, immunological and prognostic features. Eur J Cancer 27: I4 16, I99 1

1 I . Shibata D, Weiss LM, Nathwani BN, Brynes RK, Levine AM: Epstein-Barr virus in benign lymph node biopsies from individuals infected with the human immunodeficiency virus is associated with the concurrent or subsequent development of non-Hodgkin’s lym- phoma. Blood 77:1527, 1991

12. The Non-Hodgkin’s Lymphoma Pathologic Classification Project: National Cancer Institute Sponsored Study of Non-Hodgkin’s Lymphomas: Summary and description of a working formulation for clinical usage. Cancer 49:2 I 12, 1982

13. Sixbey JW, Shirley P, Chesney PJ, Buntin DM, Resnick L Detection of a second widespread strain of Epstein-Barr virus. Lancet 2:761, 1989

14. Shibata D, Weiss LM: Epstein-Barr virus-associated gastric adenocarcinoma. Am J Pathol 140:769, 1992

15. Sambrook J, Fritsch EF, Maniatis T: Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY, Cold Spring Harbor Laboratory, 1989, p 9.31

16. Raab-Traub N, Flynn K: The structure of the termini of the Epstein-Barr virus as a marker of clonal cellular proliferation. Cell 47:883, 1986

17. Sample J, Brooks L, Sample C, Young L, Rowe M, Gregory C, Rickinson A, Kieff E: Restricted Epstein-Barr virus protein expression in Burkitt lymphoma is due to a different Epstein-Barr nuclear antigen I transcription initiation site. Proc Natl Acad Sci USA 88:6343, 199 I

18. Kawasaki ES: Amplification of RNA, in Innis MA, Gelfand DH, Sninsky JJ, White TJ (eds): PCR Protocols: A Guide to Methods and Applications. San Diego, CA, Academic, 1990, p 2 1

19. Brooks L, Yao QY, Rickinson AB, Young LS: Epstein-Barr virus latent gene transcription in nasopharyngeal carcinoma cells: Coexpression of EBNA I , LMPI, and LMP2 transcripts. J Virol 66: 2689, 1992

20. Howe JG, Steitz JA: Localization of Epstein-Barr virus-en- coded small RNAs by in situ hybridization. Proc Natl Acad Sci USA 83:9006, 1986

21. Young LS. Yao QY, Rooney CM, Sculley TB, Moss DJ, Ru- piani H, Laux G, Bomkamm GW, Rickinson AB: New type B isolates of Epstein-Barr virus from Burkitt’s lymphoma and from normal individuals in endemic areas. J Gen Virol 68;2853, 1987

22. Guarner J, del Rio C, Carr D, Hendrix LE, Eley JW, Unger ER: Non-Hodgkin’s lymphomas in patients with human immuno- deficiency virus infection. Cancer 68:246O, 199 I

23. MacMahon EM, Glass JD, Hayward SD, Mann RB, Becker PS, Charache P, McArthur JC, Ambinder R F Epstein-Barr virus in AIDS-related primary central nervous system lymphoma. Lancet 338: 969, 1991

24. Shiramizu B, Herndier B, Meeker T, Kaplan L, McGrath M: Molecular and immunophenotypic characterization of AIDS-asso- ciated, Epstein-Barr virus-negative, polyclonal lymphoma. J Clin Oncol 10:383, 1992

25. Epstein MA, Achong BG: The relationship of the virus to Burkitt’s lymphoma, in Epstein MA. Achong BG (eds): The Epstein- Barr Virus. New York, NY, Springer-Verlag, 1979, p 321

26. Gutierrez M, Bhatia K, Barriga F, Diez B, Sackmann Murial F, de Andreas ML, Epelman S, Risueno C, Magrath I T Molecular epidemiology of Burkitt’s lymphoma from South America: Differ- ences in breakpoint location and Epstein-Barr virus association from tumors in other world regions. Blood 79:326 I , 1992

27. Knowles DM, Chamulak GA, Subar M, Burke JS, Dugan M, Wernz J, Slywotzky C, Pelicci PG, Dalla-Favera R, Raphael B: Lym- phoid neoplasia associated with the acquired immunodeficiency syn- drome (AIDS). Ann Intern Med 108:744, 1988

28. Henderson E, Miller G, Robinson J, Heston L: Efficiency of transformation of lymphocytes by Epstein-Barr virus. Virology 76: 152, 1977

EBV-ASSOCIATED HIV-RELATED LYMPHOMA 2109

29. Starzl TE, Porter KA, Iwatsuki S, Rosenthal JT, Shaw JBW, Atchison RW, Nalesnik MA, Ho M, Griffith BP, Hardesty RL, Jaffe R, Bahnson HT: Reversibility of lymphomas and lymphoproliferative lesions developing under cyclosporin-steroid therapy. Lancet 1:583, 1984

30. Cleary ML, Sklar J: Lymphoproliferative disorders in cardiac transplant recipients are multiclonal lymphomas. Lancet 2:489, 1984

3 1. Cleary ML, Nalesnik MA, Shearer WT, Sklar J Clonal analysis of transplant-associated lymphoproliferations based on the structure of the genomic termini ofthe Epstein-Barr virus. Blood 72:349, 1988

32. Hanto WD, Grizzera G, Gajl-Peczalska KJ, Simmons R L Epstein-Barr virus, immunodeficiency, and B cell lymphoprolifera- tion. Transplantation 39:347, 1985

33. Neri A, Barriga F, Inghirami G, Knowles DM, Neequaye J, Magrath IT, Dalla-Favera R: Epstein-Barr virus infection precedes clonal expansion in Burkitt’s and acquired immunodeficiency syn- drome-associated lymphoma. Blood 77: 1092, 199 1

34. Katz BZ, Raab-Traub N, Miller G: Latent and replicating forms of Epstein-Barr virus DNA in lymphomas and lymphoprolif- erative disease. J Infect Dis 160:589, 1989

35. Pelicci P-G, Knowles DM, Arlin ZA, Wieczorek R, Luciw P, Dina D, Basilico C, Dalla-Favera R: Multiple monoclonal B cell expansions and c-myc oncogene rearrangements in acquired immune deficiency syndrome-related lymphoproliferative disorders. J Exp Med 164:2049, 1986

36. Young L, Alfieri C, Hennessy K, Evans H, OHara C, Anderson C, R i g J, Shapiro RS, Rickinson A, Kieff E, Cohen JI: Expression of Epstein-Barr virus transformation-associated genes in tissues of patients with EBV lymphoproliferative disease. N Engl J Med 321: 1080, 1989

37. Thomas JA, Hotchin NA, Allday MJ, Amlot P, Rose M, Ya- coub M, Crawford DH: Immunohistology of Epstein-Barr virus-as- sociated antigens in B cell disorders from immunocompromised in- dividuals. Transplantation 49:944, 1990

38. Rowe M, Young LS, Crocker J, Stokes H, Henderson S, Rick- inson A B Epstein-Barr virus (EBV)-associated lymphoproliferative disease in the SCID mouse model: Implications for the pathogenesis of EBV-positive lymphomas in man. J Exp Med 173: 147, 199 I

39. Pisa P, Cannon MJ, Pisa EK, Cooper NR, Fox RI: Epstein- Barr virus induced lymphoproliferative tumors in severe combined immunodeficient mice are oligoclonal. Blood 79: 173, 1992

40. Schaefer BC, Woisetschlaeger M, Strominger JL, Speck SH: Exclusive expression of Epstein-Barr virus nuclear antigen 1 in Burkitt lymphoma arises from a third promotor, distinct from the promotors used in latently infected lymphocytes. Proc Natl Acad Sci USA 88: 6550, 1991

41. Knowles DM, Inghirami G, Ubriaco A, Dalla-Favera R: Mo- lecular genetic analysis of three AIDS-associated neoplasms of un- certain lineage demonstrates their B-cell derivation and the possible pathogenetic role of the Epstein-Barr virus. Blood 73:792, 1989

42. Pallesen G, Hamilton-Dutoit SJ, Rowe M, Lisse I, Ralfkiaer E, Sandveu K, Young LS: Expression of Epstein-Barr virus replicative proteins in AIDS-related nomHodgkin’s lymphoma cells. J Pathol 165:289, 1991

43. Rowe M, Lear AL, Groom-Carter D, Davies AH, Rickinson AB: Three pathways of Epstein-Barr virus gene activation from EBNAI-positive latency in B lymphocytes. J Virol66: 122, 1992

44. Herbst H, Dallenbach F, Hummel M, Niedobitek G, Pileri S, Muller-Lantzsch N, Stein H: Epstein-Barr virus latent membrane protein expression in Hodgkin and Reed-Sternberg cells. Proc Natl Acad Sci USA 88:4766, 1991

45. Pallesen G, Hamilton-Dutoit SJ, Rowe M, Young LS: Expres- sion of Epstein-Barr virus latent gene products in tumor cells of Hodgkin’s disease. Lancet 337:320, 199 1

46. Wang D, Leibowitz D, Kieff E: An EBV membrane protein expressed in immortalized lymphocytes transforms established rodent cells. Cell 43931, 1985

47. Wang D, Leibowitz D, Wang F, Gregory C, Rickinson A, Larson C, Springer T, Kieff E Epstein-Barr virus latent infection membrane protein alters the human B-lymphocyte phenotype: Dele- tion of the amino terminus abolishes activity. J Virol62:4173, I988

48. Murray RJ, Wang D, Young LS, et al: Epstein-Barr virus- specific cytotoxic T-cell recognition of transfectants expressing the virus-coded latent membrane protein LMP. J Virol 62:3747, 1988

49. Rickinson AB, Young LS, Rowe M: Influence of the Epstein- Barr virus nuclear antigen EBNA 2 on the growth phenotype of virus transformed B cells. J Virol 6 I: I3 IO, 1987

50. Yarchoan R, Redfield RR, Broder S: Mechanisms of B cell activation in patients with acquired immunodeficiency syndrome and related disorders. J Clin Invest 78:439, 1986

5 I. Martinez-Maza 0, Crabb E, Mitsuyasu RT, Fahey JL, Giorgi JV: Infection with the human immunodeficiency virus (HIV) is as- sociated with an in vivo increase in B lymphocyte activation and immaturity. J Immunol 138:3720, 1987

52. Haluska FG, Russo G, Kant J, Andreef M, Croce CM: Mo- lecular resemblance of an AIDS-associated lymphoma and endemic Burkitt lymphomas: Implications for their pathogenesis. Proc Natl Acad Sci USA 863907, 1989

53. Whittle HC, Brown J, Marsh K, Greenwood BM, Seidelin P, Tighe H, Wedderbum L T-cell control of Epstein-Barr virusinfected B-cells is lost during P. falciparum malaria. Nature 312:449, 1984

54. OConner G T Persistent immunological stimulation as a factor in oncogenesis with special reference to Burkitt’s tumor. Am J Med 48:279, 1970

55. Gaidano G , Ballerini P, Gong JZ, Inghirami G, Neri A, New- comb EW, Magrath IT, Knowles DM, Dalla-Favera R: p53 mutations in human lymphoid malignancies: Association with Burkitt lym- phoma and chronic lymphocytic leukemia. Proc Natl Acad Sci USA 88:5413, 1991