Abstract from "Lymphoma... The Next Questions" Conference,October 21-23, Palm Desert, CA
Special thanks to the MD Anderson Cancer Center and Imedex USA

Part I
Part II

Part I

Significant advances in molecular biology have provided major insights into the mechanisms that control lymphoma development.  In contrast to epithelial malignancies that are complicated by tumor cell heterogeneity and microsatellite instability, non-Hodgkin's lymphomas (NHL) are more predictably clonal and are associated with well-characterized molecular defects that underlie their pathogenesis.  Examples include the bcl-l translocation characteristic of mantle cell lymphoma (MCL) [t(11;14)(q13;q32)], the bcl-2 translocation associated with follicular lymphoma (FL) [t(14;18)(q32;q21)], the bcl-3 translocation found in small lymphocytic lymphoma (SLL)/chronic lymphocytic leukemia (CLL) [t(14;19)(q32;q13)], bcl-6 rearrangements associated with diffuse large B cell lymphomas (DLBC) and some FL involving chromosomal breakpoints at 3q27, c-myc rearrangements found characteristically in Burkitt's lymphoma [t(8;14)(q24;q32)], pax-5 translocations seen in lymphoplasmacytic lymphoma (LPL) [t(9;14)(pl 3;q32)], bcl-10 rearrangements described recently in low-grade B cell lymphomas of mucosa-associated lymphoid tissue (MALT)-type [t(1;14)(p22;q32)] and the newly characterized cytogenetic defect seen in MALT lymphomas from a variety of anatomical sites, AP12-MLT, associated with t(11;18)(q21;q21).  The immunoglobulin heavy chain locus is frequently involved as a partner in these balanced translocations, a finding in concert with the B cell nature of these tumors.  Knowledge of the molecular abnormalities or karyotypic defects present within lymphoma cells provides important insights into the pathogenesis of the lymphoid neoplasm, but does not directly correlate with the phenotype and thus may not be helpful in predicting the clinical course and eventual outcome of patients with these diseases.  In contrast, the phenotype of the malignant cells may have a significant effect on the clinical behavior of the lymphoma, and thus be prognostically important.  Understanding the differences between molecular genetic events and the phenotypic expression of a functional protein are the key elements in resolving both the pathogenesis and the determination of the important biologic correlates that underlie the clinical behavior of the NHLS.

Approximately 85-90% of FLs are characterized by a non-random cytogenetic abnormality, the t(14,18)(q32;q21).  This balanced translocation deregulates expression of a normal gene on chromosome 18q21, known as bcl-2.  This gene codes for a 26kD protein found within the nuclear membrane and endoplasmic reticulum that acts by as yet unknown mechanism to inhibit programmed cell death (apoptosis).  The critical event in FL lymphomagenesis is the constituent over-expression of the bcl-2 gene, leading to prolonged survival of a population of germinal center cells normally destined to die via apoptosis.  It is of interest that not all FL cases have this translocation.  In fact, approximately 10-15% demonstrate a clonal karyotype other than t(14;18).  When the molecular/cytogenetic information is combined with Bcl-2 protein expression data, several conclusions can be drawn.  Not all FL cases with a bcl-2 gene rearrangement or t(14;18) cytogenetic abnormality produce a functional Bcl-2 molecule.  The precise mechanism that allows the cells to escape apoptosis is unknown, but may involve the timely over-expression of another member of the bcl-2 gene family.  Alternatively, one third of the FL cases that are karyotypically clonal, but lack a t(14;18), over-express the Bcl-2 molecule, and do so by a mechanism other than translocation.  Ultimately, all cases of FL that express Bcl-2 may share a common clinical course that is in contrast to the behavior of Bcl-2 protein-negative FL.  In our experience, Bcl-2 protein-negative FL have superior overall survival.  In the case of FL, molecular cytogenetic events are closely linked to phenotype, although the relationship is not precise.  Most cases of FL with a t(14;18) over-express the Bcl-2 protein.  A clear understanding of the divergence between molecular genetics and phenotype is seen when one examines the diffuse small B cell NHLs (MCL, SLUCLL, LPL, MALT lymphomas).  For the most part, the t(14;18) cytogenetic abnormality or bcl-2 gene rearrangement is not present in these lymphomas, yet the vast majority of neoplastic cells in these NHLs over-express Bcl-2 protein by a mechanism other than translocation.  Therefore, despite having a completely different pathogenesis, these low-grade B cell NHLs share a common Bcl-2 phenotype that likely plays a role in determining the ultimate clinical course of these tumors.

De novo DLBC lymphoma will harbor a t(14;18)(q32;q21) or the molecular equivalent, a bcl-2 rearrangement, in approximately 15-20% of cases.  These data suggest that some DLBC lymphomas share a relationship with FL, and may have arisen out of a clinically silent low-grade lymphoma background.  Alternatively, a germinal center cell with a t(14;18) translocation that was destined for the follicle may have suffered another mutational event(s) in rapid succession, resulting in a diffuse large cell histology.  The precise relationship between FLs and DLBC lymphomas with a t(14;18) remains unresolved.

Bcl-2 protein is expressed in approximately 45-60% of DLBC lymphomas and is unrelated to the presence of a bcl-2 oncogene rearrangement.  Expression of Bcl-2 is much more common in nodal DLBC lymphomas as compared to those of extranodal origin.  Gene amplification has been shown to be a mechanism of over-expression, and more recently mutations in the open reading frame have also been demonstrated to have a role.  The majority of the published studies have shown that the presence of a bcl-2 rearrangement has no prognostic impact in DLBC lymphoma, but expression of the Bcl-2 protein has a significant effect on both overall and failure-free survival.  Bcl-2 expression confers a relative chemo-resistance on the tumor cells by inhibiting an important apoptotic pathway within the cell.  This translates into a continuous risk of late relapse and diminished survival, but does not affect the complete remission rate.

The bcl-6 gene located on chromosome 3q27 codes for a DNA-binding transcriptional repressor that plays an important role in the development of both the normal germinal center and T cell-dependent antigen responses.  This 95 kD nuclear phosphoprotein is normally expressed in the germinal center and in a small number of interfollicular T cells.  Constitutive expression of Bcl-6 may be involved with phenotypic changes in germinal center cells by affecting differentiation and/or apoptosis.  When normal germinal center B cells survive the process of affinity maturation and leave the germinal center environment to become either plasma cells or long-lived memory B cells, they normally down-regulate Bcl-6 expression.  It has been hypothesized that a block in normal down-regulation of Bcl-6 may cause genetic instability in the germinal center cells, ultimately leading to neoplastic transformation.

The bcl-6 gene was discovered by virtue of its involvement in a balanced translocation involving the immunoglobulin (1g) heavy chain gene on chromosome 14q32.  The most common cytogenetic abnormality is the t(3;1 4)(q27;q32), but t(2;3)(p12;q27) and t(3;22)(q27;q11) involving the light chain genes are also seen.  In contrast to other oncogene rearrangements involving the lg loci, bcl-6 rearrangements are promiscuous and involve a number of different partners throughout the genome.  The resultant bcl-6 over-expression is thought to involve a novel mechanism of promoter substitution, whereby ectopic promoters are translocated into the proximity of the coding exons and up-regulate transcription of the gene.  Furthermore, 5'mutations in the non-coding region of the bcl-6 gene are common in both DLBC lymphomas (80%) and FLs (50%).  These are believed to result from ectopic activity of the lg variable region gene hypermutation mechanism, and serve as a signature of passage through the germinal center.  They are not believed to play a direct role in the pathogenesis of NHLS.

Translocations involving bcl-6 are seen in approximately 35-40% of DLBC lymphomas and a smaller percentage of FLs (15 -20%).  They can be detected using either Southern blot or classical cytogenetics, but importantly these two tests are not mutually exclusive.  The bcl-6 rearrangements that involve the Ig loci can be detected using Southern blot in 80% of cases using the Sac 4.0 probe, but the non-ig translocations involving 3q27 are detected at a much lower frequency  (approximately 25%).  Similarly, about 30-40% of cases with a bcl-6 rearrangement using Southern blot analysis reveal no demonstrable abnormality of chromosome 3 using classical cytogenetics.  These cases are considered cryptic or “cytogenetically silent”.  Cumulatively, these data demonstrate that in order to accurately identify all cases and reduce the number of false negatives, both techniques are required.  Thus, studies designed to determine the prognostic impact of bcl-6 gene rearrangements in either DLBC lymphoma or FL should be viewed with skepticism if both classical cytogenetics and Southern blot analysis are not included in the study design.

Bcl-6 protein expression, like expression of Bcl-2 in DLBC lymphoma, is unrelated to the presence of the translocation.  The vast majority of FLs express Bcl-6 protein, but the translocation is found in only 5-10% of cases.  Most cases of SLUCLL, MCL, LPL and MALT lymphomas do not express Bcl-6 protein.  The expression of Bcl-6 is seen in the majority of DLBC lymphomas, but is much less frequently seen in immunoblastic lymphomas as compared with large non-cleaved cell lymphomas.  This reflects the fact that Bcl-6 expression serves as a surrogate marker of a germinal center cell origin.  Together with CD138 (syndecan), cases of DLBC lymphoma can be assigned to germinal center stage of differentiation (Bcl-6+, CD138?) vs a post-germinal center stage (Bcl-6?, CD138+).  The clinical impact of these different phenotypes has not been determined.  Similarly, the prognostic significance of Bcl-6 protein expression in DLBC lymphoma is untested.

In summary, molecular/cytogenetic events are important for understanding the pathogenesis and perhaps etiology of NHLS, but do not allow us to predict the functional phenotype of the tumor cells.  There appear to be several distinct mechanisms by which neoplastic cells can up-regulate expression of critical proteins, the net result of which will be important in determining the clinical behavior of the tumor.

References:

1. Horsman DE, Gascoyne RD, Coupland RW, et al.  Comparison of cytogenetic analysis, southern analysis. and polymerase chain reaction for the detection of t(14;18) in follicular lymphoma.  Am J Clin Pathol 103:472,1995

2. Horsman DE, McNeil BK, Anderson M, et al. Frequent association of t(3;14) or variant with other lymphoma-specific translocations. Br J Haematol 89:569, 1995

3. Fan H, Gulley ML, Gascoyne RD, et al.  Molecular methods for detecting t(11;14) translocations in mantle-cell lymphomas.  Diagn Mol Pathol 7:209,1998

4. Michaux L, Dieriamm J, Wlodarska 1, et al. t(14;19)/BCL3 rearrangements in lymphoproliferative disorders: a review of 23 cases.  Cancer Genet Cytogenet 94:36,1997

5. Willis TG, Jadayel DM, Du MQ et al.  Bcll 0 is involved in t(1;14)(p22;q32) of MALT B cell lymphoma and mutated in multiple tumor types.  Cell 96:35,1999

6. Dierlamm J, Baens M, Wlodarska 1, et al.  The apoptosis inhibitor gene AP12 and a novel 18q gene, MLT, are recurrently rearranged in the t(11;18)(q21q21) associated with mucosa- associated lymphoid tissue lymphomas [In Process Citation].  Blood 93:3601,1999

7. Tsujimoto Y, Cossman J, Jaffe E, Croce CM.  Involvement of the bcl-2 gene in human follicular lymphoma.  Science 228:1440,1985

8. Skinnider BF, Horsman DE, Dupuis B, Gascoyne RD.  Bci-6 and Bcl-2 protein expression in diffuse large B-cell lymphoma and follicular lymphoma: correlation with 3q27 and 18q21 chromosomal abnormalities [In Process Citation].  Hum Pathol 30:803,1999

9. Hill ME, MacLennan KA, Cunningham DC, et al.  Prognostic significance of BCL-2 expression and bcl-2 major breakpoint region rearrangement in diffuse large cell non-Hodgkin's lymphoma: a British National Lymphoma Investigation Study.  Blood 88:1046,1996

10. Gascoyne RD, Adomat SA, Krajewski S, et al.  Prognostic significance of Bcl-2 protein expression and Bcl-2 gene rearrangement in diffuse aggressive non-Hodgkin's lymphoma.  Blood 90:244,1997

11. Monni 0, Joensuu H, Franssila K, et al. BCL2 overexpression associated with chromosomal amplification in diffuse large B-cell lymphoma. Blood 90:1168,1997

12. Monni 0, Franssila K, Joensuu H, Knuutila S. BCL2 overexpression in diffuse large B-cell lymphoma.  Leuk Lymphoma 34:45,1999

13. Dalla-Favera R, Ye BH, Cattorefti G, et al.  BCL-6 in diffuse large-cell lymphomas.   Important Advances in Oncology: 139, 1996

14. Carbone A, Gaidano G, Gloghini A, et al.  Differential expression of BCL-6, CD138/syndecan-1, and Epstein-Barr virus-encoded latent membrane protein-1 identifies distinct histogenetic subsets of acquired immunodeficiency syndrome-related non-Hodgkin's lymphomas. Blood 91:747,1998

15. Offit @ Lo Coco F, Louie DC, et al.  Rearrangement of the bcl-6 gene as a prognostic marker in diffuse large-cell lymphoma.  New England Journal of Medicine 331:74,1994

16. Shen HM, Peters A, Baron B, et al.  Mutation of BCL-6 gene in normal B cells by the process of somatic hypermutation of lg genes.  Science 280:1750,1998

17. Raible MD, Hsi ED, Alkan S. Bcl-6 protein expression by follicle center lymphomas.  A marker for differentiating follicle center lymphomas from other low-grade lymphoproliferative disorders.  Am J Clin Pathol 11 2:101, 1999
 

Part II

In virtually all patients with a suspected hemato-lymphoid neoplasm, an open biopsy is required in order to establish a definitive diagnosis.  The biopsy should be done in consultation with a hematopathologist, ensuring that fresh material is submitted for lymphoma protocol studies including immunophenotypic analysis using flow cytometry, molecular genetic analysis and possible cytogenetic studies.  In academic centers, this approach facilitates accurate diagnoses and maximizes the use of this valuable resource for clinical research.  For those patients where there is a high clinical index of suspicion of a diagnosis of lymphoma, an open biopsy should be performed following the protocol outlined above. Importantly, fine needle aspiration and needle core biopsy techniques should not be a consideration as a diagnostic strategy unless co-morbid disease or surgical inaccessibility precludes such an approach.  However, fine needle aspiration and/or needle corebiopsy techniques do have a role in staging of other sites of disease in patients with an established diagnosis, as a method for procuring additional cells for molecular or cytogenetic studies prior to therapy and in patients with suspected relapse.  It is important to remember that the diagnosis of lymphoma in most cases is based upon the recognition of altered architecture and is not made using cytologic criteria.

In the majority of cases, good quality haematoxylin-eosin slides together with paraffin section immunostains will provide an accurate diagnosis, as routine histologic studies remain the gold standard for diagnosis.  Detailed subclassification is now possible using this approach because of the growing list of paraffin-reactive antibody reagents.  This list includes antibodies recognizing CD5, CD10, CD23, Bcl-1, Bcl-2, CD21, CD43 and ALK to name a few.  Furthermore, formalin fixation can be used for molecular genetic studies using polymerase chain reaction (PCR) techniques if required.  These studies provide useful data for establishing clonality, resolving lineage and determining the presence or absence of specific translocations helpful for the subclassification of the small B cell lymphomas.  Precise classification of lymphoid neoplasms today often requires input from a variety of sources including morphology, immunophenotype, molecular genetics and cytogenetics.  This approach to diagnosis is similar to that adopted by our acute leukemia colleagues (MIC = morphology, immunophenotype and cytogenetics), and recognizes that distinct clinicopathologic entities have a characteristic morphology, together with a predictable immunophenotype and molecular genetic/cytogenetic correlate.  In difficult cases or those with borderline histology, a multiparameter approach is frequently required to arrive at a confident diagnosis.  This approach has particular merit when dealing with small biopsies of extranodal sites such as endoscopic gastric biopsies or 4 mm punch biopsies of skin.  For example, if a gastric biopsy is small and the histologic findings fall short of a diagnosis of mucosa associate lymphoid tissue (MALT) lymphoma, then two options are available.  Firstly, paraffin sections can be stained for K and ? in order to establish immunophenotypic clonality and/or DNA can be extracted from the block for PCR studies to determine B cell clonality.  If these fail to produce convincing evidence of lymphoma, a repeat biopsy can be performed to include in addition to routine histology, fresh material submitted for molecular genetic studies.  Importantly, phenotypic or molecular evidence of clonality is of limited use if there is insufficient material to establish the required histologic diagnosis and subclassification needed for treatment planning.  Specific cases will be discussed in this session in order to highlight these points, with emphasis on a multiparameter approach to diagnosis.

References:

1. Jaffe ES.  Surgical pathology of the lymph nodes and related organs, vol. 16(ed2nd).  Philadelphia, W. B. Saunders Company, 1995

2. Harris NL, Jaffe ES, Stein H, et al.  A revised European-American classification of lymphoid neoplasms: a proposal from the International Lymphoma Study Group.  Blood 84:1361,1994

3. Ashton-Key M, Diss TC, Isaacson PG, Smith ME.  A comparative study ofthe value of immunohistochemistry and the polymerase chain reaction in the diagnosis offollicular lymphoma.  Histopathology 27:501,1995

4. Cossman J, Uppenkamp M, Sundeen J, et al.  Molecular genetics and the diagnosis of lymphoma.  Archives of Pathology & Laboratory Medicine 112:117,1988

5. Algara P, Martinez P, Sanchez I, et al.  The detection of B-cell monoclonal populations by polymerase chain reaction: accuracy ofapproach and application in gastric endoscopic biopsy specimens.  Human Pathology 24:1184,1993

6. Algara P, Soria C, Martinez P, et al.  Value of PCR detection ofTCR gamma gene rearrangement in the diagnosis of cutaneous lymphocytic infiltrates.  Diagnostic Molecular Pathology 3:275,1994

7. Ben-Yehuda D, Polliack A, Okon E, et al. image-guided core-needle biopsy in malignant lymphoma: experience with 100 patients that suggests the technique is reliable.  Journal of Clinical Oncology 14:2431, 1996

8. Chhanabhai M, Adomat SA, Gascoyne RD, et al.  Clinical utility of heteroduplex analysis ofTCR gamma gene rearrangements in the diagnosis of T-cell lymphoproliferative disorders.  Am J Clin Pathol 108:295, 1997

9. Gascoyne RD.  Establishing the diagnosis of lymphoma: from initial biopsy to clinical staging.  Oncology (Huntingt) 12:11, 1998

10. Jennings CD, Foon KA.  Recent advances in flow cytometry: application to the diagnosis of hematologic malignancy.  Blood 90:2863,1997

11. Katz RL Controversy in fine-needle aspiration of lymph nodes.  A territorial imperative?  American Journal of Clinical Pathology 108:S3, 1997

12. Horsman DE, Gascoyne RD, Coupland RW, et al.  Comparison of cytogenetic analysis, southern analysis, and polymerase chain reaction for the detection of t(14;18) in follicular lymphoma.  Am J Clin Pathol 103:472,1995