1349 Clonality Analysis of B-Lymphoid Proliferations Using the Polymerase Chain Reaction Maria D. Lozano, M.D.,Ph.D. Anne Tierens, M.D. Timothy C. Greiner, M.D. Robert S. Wickert, M.S. Dennis D. Weisenburger, M.D. Wing C. Chan, M.D. Department of Pathology and Microbiology University of Nebraska Medical Center, Omaha, Nebraska. BACKGROUND. Polymerase chain reaction (PCR) based assays are becoming more reliable, simpler, and faster alternatives to traditional Southern blot hybridization (SBH) analysis for the detection of clonal immunoglobulin heavy chain (IgH) gene rearrangements. However, a variety of technical approaches have been reported with markedly different results. METHODS. We analyzed the frozen tissue of 147 neoplastic and hyperplastic lesions on which SBH had previously been performed. Semi-nested and single-step PCR methods were compared. Consensus primers to the joining segments and the framework region (FR) 111 of the variable segments of the IgH gene were used. All PCR products were analyzed by polyacrylamide gel electrophoresis (PAGE). Thirteen samples were re-analyzed using a denaturing gradient gel electrophoresis (DGGE) system. RESULTS. The overall concordance between SBH and semi-nested PCR assays was 80.2%. In the non-Hodgkin’s lymphoma (NHL) group, 75% of the cases with IgH rearrangements by SBH were found to be monoclonal by PCR. Regardless of type of lesion, 71.7% of the cases with IgH rearrangements by SBH were found to be clonal by PCR. The concordance between the semi-nested and single-step procedures was 87.1%. DGGE was helpful in clarifying the results for cases in which the PAGE analysis was difficult to interpret. CONCLUSIONS. PCR analysis of IgH gene rearrangements was found to be an efficient technique for the initial determination of clonality in lymphoid proliferations. The single-step method had an advantage over the semi-nested method because of its simplicity and speed. The DGGE system was useful for the assessment of clonality in cases with equivocal results after PAGE. However, a combination of these techniques in specific cases may achieve higher specificity and sensitivity. Cancer 1996; 77:1349-55. 0 1996 American Cancer Society. KEYWORDS polymerase chain reaction, non-Hodgkin’s lymphoma, immunoglobulin gene rearrangement, clonality. Supported in part by USPHS Grant CA-36727 from the National Cancer Institute, Department of Health and Human Services. The authors thank Debra Lytle for technical assistance, George Pallas for photography, and Karen Hansen for secretarial assistance. Address for correspondence: Wing C. Chan, M.D., Department of Pathology and Microbiology, University of Nebraska Medical Center, 600 South 42nd Street Omaha, NE 68198-3135. Received May 4,1995; revision received August 4, 1995; accepted August 4, 1995. 0 1996 American Cancer Society T he determination of clonality plays an important role in the study of lymphoproliferative disorders. Recently, the polymerase chain reaction (PCR) has been increasingly employed for the rapid detection of clonal antigen receptor gene rearrangements in lymphoid proliferations, and is being used as an alternative to traditional Southern blot hybridization (SBH) analysis. The genes encoding immunoglobulin molecules undergo an ordered process of rearrangement during B-lymphocyte development. During the course of the rearrangement, various gene segments, coding for the immunoglobulin heavy chain (IgH), which are widely separated in their germline configuration, are brought into proximity. The result of this process is an enormous variability of the IgH variable region generated by the rearrangement of variable (VH), diversity (DH), and joining (JH) 1350 CANCER April 1,1996 / Volume 77 / Number 7 region The variability of the IgH gene is concentrated in three regions known as complementary determining regions (CDR)I, 11, and 111, which are separated by four framework regions (FR) exhibiting a lesser degree of variability. Among them, CDR I11 is generated by the largely random recombination of the different VH, DH, and JH segments, resulting in marked combinatorial diversity. In addition, sequence heterogeneity of CDR I11 is increased by the deletion of some nucleotides and the insertion of other ones at the 3’ end of the joining sites by the enzyme terminal deoxynucleotidyl transferase (TdT).4 This region, therefore, can be exploited as a clonal marker in DNA amplification assays since the unique sequences derived from the rearrangement process are passed on to subsequent generations of Most of the reported PCR studies of clonality in Blymphoid proliferations are based on “consensus” primers which hybridize with conserved sequences present in most of the V and J segments of the IgH gene. However, the number of different technical approaches taken has resulted in marked differences (55%-96% sensitivity) in the reported result^.^-^^ To test the sensitivity and specificity of PCR as a method for the determination of clonality in B cell lymphoproliferative disorders, we have evaluated the performance characteristics of two different PCR methods for the detection of IgH gene rearrangements. We also evaluated the usefulness of denaturing gradient gel electrophoresis (DGGE) in the study of some equivocal cases to further enhance the reliability of PCR analysis. MATERIAL AND METHODS Samples We analyzed the frozen tissue of 147 fully-characterized hematolymphoid proliferations on which SBH had been previously performed. All samples were obtained from the files of the Nebraska Lymphoma Study Group Registry. These included 96 patients with non-Hodgkin’s lymphoma (NHL),including 88 B cell and 8 T cell NHLs by immunohistochemistry; 15 atypical lymphoid proliferation (ALP); 17 non-neoplastic tissue, including 6 with reactive hyperplasia (RH), 5 with follicular hyperplasia (FH),3 with angiofollicular hyperplasia (AFH),1with tonsillitis, and 2 normal lymph nodes (NL); and 19 miscellaneous, including 8 cases of Hodgkin’sdisease (HD),8 post-transplant lymphoproliferative disorders (PTLD),2 cases of acute lymphoblastic leukemia (ALL), and 1 chronic myeloid leukemia (CML). Stored frozen DNA used previously for SBH was used for PCR analysis for all of the patients. DNA Extraction DNA was extracted from the frozen tissue as previously de~cribed.2~ Briefly, the samples were homogenized in a solution containing 10 mM EDTA, 75 mM NaCl at pH 8.0, TABLE 1 Sequences of Oligonucleotide Primers V670 ]H external JH nested JH nested (GC-clamped) ]H single CTG TCG ACA CGG CCG TGT A’IT ACT G ACCTGAGGAGACGGTGACC ACCAGGGTCCCTTGGCCCCA 5’CGCCCGCCGCGCCCCGCGCCCGGCCCG CCGCCCCCGCCCGACCAGGGTCCCTTG GCC CCA G 3’ AACTGCAGAGGAGACGGTGACC 0.1 volume of 10% SDS, and 0.15 mg/mL of proteinase K. Two extractions with phenol: chloroform: isoamyl alcohol (25:24:1),followed by one extraction with chloroform: isoamy1 alcohol (24:1),were done to remove contaminants. The DNA was then precipitated with 0.1 volume of 5M ammonium acetate and 5 volumes of cold (-20 “C) 80% ethanol, resuspended in TE buffer (10 mM Tris-HC1, 1 mM EDTA), and quantitated spectrophotometricaly. Oligonucleotide Primers and PCR Amplification The primer sequences used are shown in Table 1. All samples were analyzed by a semi-nested method based on the technique of Wan and colleagues.9In the first PCR round, a “consensus” primer to the third framework (FR 111) region of the VH genes was used with a consensus JH primer. In the nested round, the same VH primer was used with a consensus JH primer, internal to the JH primer used for the first round. All oligonucleotide primers were synthesized by Oligo’s Etc. [Wilsonville, OR). Initially, 1 pg of DNA was denatured at 94 ”C for 5 minutes in 1 X PCR buffer (50 mM KCl, 10 mM Tris-HC1 at pH 8.3, 0.1 mglml gelatin), with 2.5 mM M g Q , 0.45% Nonidet P40 and 0.45% Tween 20 in a volume of 25 pl, overlaid with 40 PI mineral oil. While still at 94 “C, the addition of 25 p1 to a master mix brought the final reaction concentrations to: 1 X PCR Buffer, 2.0 mM MgCI,, 0.1 mM dNTP, 0.8 pM of each primer, and 1 Unit Taq polymerase per reaction. To maximize the yield of a specific product, the Taq polymerase was not added to the master mix until it reached 94 A PCR cycle consisted of denaturation for 40 seconds at 94 “C, annealing for 40 seconds at 55 “C, and extension for 40 seconds at 72 “C. The first round of PCR was performed for 30 cycles. The second round consisted of 20 cycles based on the previous parameters. In the second round, 5 pl of a 1:500 dilution of the initial products was used as a template. In this semi-nested reaction, nonspecific amplification products of the first round were not amplified in the second round, thereby enhancing the specificity of the entire reaction. Positive controls consisted of DNA extracted from the Namalwa B cell line diluted to lo-’ and lo-‘ in O C Z 6 Detection of IgH Rearrangements by PCR/Lozano et al. 1351 TABLE 2 Comparison of SBH and Semi-Nested PCR Regardless of the Type of Lymphoid Proliferation Semi-nested IgH-PCR Clonal Non-clonal Clonal' 84 (100%) 60 (71.5%) 24 (28.5%) Non-dona!** 63 (100%) 5 (8%) 58 (92%) SBH FIGURE 1. Same cases analyzed by single-step PCR (right) and seminested PCR (left). Lanes A, B, and C represent three patients with clonal bands (arrows). In one patient (Lane B), two bands are seen in the seminested procedure but only one is seen in the single-step PCR assay. Polyclonal cases appear in both techniques as a smear in the 100 bp range (Lane 0). M: DNA size marker. FIGURE 2. Comparison between PAGE and DGGE analysis. Lane A was thought to represent bi-allelic rearrangements. In the DGGE system, only one rearranged band is seen, suggesting rearrangement of only one allele. Lane B shows a faint band in a smear. DGGE clarifies the monoclonality of this case. M: DNA size marker (not applicable to DGGE). DNA from the erythroleukemia cell line K562. Negative controls consisted of sterile water. Both controls were included in each set of reactions. All PCR reactions were performed using an Omnigene Thermocycler (Hybaid Ltd., UK). For the patients in whom there were discrepancies between SBH and PCR results, the process was repeated using the same conditions. In addition, 62 of the patients were also analyzed by a single-step PCR procedure using the same consensus VH primer and a different JH primer (JH single), as described by Sioutos and (Table 1). Initially, 0.5 pg to 1 pg of template DNA was denatured at 94 "C for 5 minutes in water in a volume of 25 pl, overlaid with 40 pI of mineral oil. While at 94 "C, the addition of 25 p1 of a master mix brought the final reaction concentration to: l'otal: 147 cases; Overall Concordance: 118 cases (80.2%) * IgH rearrangements. '* TCRB rearrangements only or germline configuration. 0.8 X PCR buffer, 0.2 mM dNTP, 1.0 mM MgC12, 0.5 pM of each primer, and 2 Units of Taq polymerase. The amplification steps consisted of 40 cycles of DNA denaturing at 94 "C for 1 minute, annealing at 59 "C for 20 seconds, and extension at 72 "C for 20 seconds. As in the seminested procedure, positive and negative controls were included in each run. For patients in whom discrepancies arose between semi-nested and single-step PCR, procedures were repeated. For both techniques, 15 p1 of the PCR products were electrophoresed in 16 cm x 20 cm x 0.75 mm, 8% polyacrylamide gels and visualized with UV light after ethidium-bromide staining. All gels were carefully analyzed following strict criteria: the presence of one or two distinct bands was considered diagnostic of monoclonality, the presence of three or more distinct bands, oligoclonality, and the presence of a smear of products, polyclonality. Thirteen cases in which none of the above patterns were unequivocally identified were also analyzed using In it, DNA fragments were separated a DGGE ~ystem.~' according to the denaturing characteristics inherent to their sequences. A GC-clamped, JH nested primer (Table 1) was used in the second amplification step of the assay. Conditions were similar to those previously described except that 5 ~1 of a 1:200 dilution of the initial products was used as a template for the second PCR. The final concentration of both primers was 0.2 pM, and the total reaction volume was 100 pl. DGGE was performed as previously described," except that an 8%polyacrylamide gel and a 40% to 70% denaturing gradient of urea formamide were used. RESULTS Monoclonality was assigned only when one or two bands in the range of 80 bp to 120 bp were clearly detected in the gel. Polyclonal and oligoclonal cases which appeared as smears or as a number of bands, were classified as non-clonal (Fig. 1). The results from 13 patients which 1352 CANCER April 1, 1996 / Volume 77 I Number 7 TABLE 3 Comparison of SBH and PCR Results among Different Types of Lymphoid Proliferations ~ IgH-PCR Southern blot NHL = 96 ALP = 15 REACTIVE = 17 MISC = 19 Clonal B Clonal T Germline Clonal B Clonal T Germline Clonal B Clonal T Germline Clonal B Clonal T Germline 68 7 21 4 1 10 2 2 13 10 2 7 Concordance* Clonal Non-clonal 51 (75%) 17 (25%) 6 (75%) 19 (90.5%) 1 2 (9.5%) 2 (50%) 0 1(lO%] 0 0 0 7 (70%) 1 0 Yes No 76 (79.2%) 20 (20.8%) 12 (80%) 3 (20%) 15 (88.3%) 2 (11.7%) 15 (79%) 4 (21%) 2 (50%) 1 9 (90%) 2 2 13 (100%) 3 (30%) 1 7 (100%) TOTAL = 147 Clonal B IgH or IgH t TCRP rearrangements; Clonal T only TCRP rearrangements; MISC Miscellaneous. ’Overall concordance of clonal and non-clonal cases in each category by SBH and PCR. were difficult to interpret in acrylamide gels were additionally analyzed in a DGGE system before final interpretation. A distinct band observed in DGGE was interpreted as evidence of monoclonality (Fig. 2 ) . Table 2 shows the results comparing SBH with the semi-nested PCR procedure. The overall concordance between SBH and PCR results was 80.2%. Regardless of the type of lymphoproliferation, 71.5% of the lesions which showed clonal IgH rearrangements by SBH were monoclonal using the semi-nested PCR technique. The false negative rate was 28.5% and the false positive rate in the whole series was 8%. The results were further analyzed according to different diagnostic groups (Table 3). In the NHL category, 75% of the lesions with IgH rearrangements by SBH were monoclonal by PCR. In this group, the percentage of false negative results was 25%. In the non-neoplastic (reactive) group, one patient with angiofollicular hyperplasia and one patient with reactive hyperplasia had IgH gene rearrangements by SBH, but were polyclonal by PCR. All patients with the germline configuration by SBH were also polyclonal by PCR. The findings in the miscellaneous group were expected to be more heterogeneous because it included PTLD and HD, however the results were found to be similar to the NHL group. In this group, three pa- tients who were non-clonal by PCR but had IgH rearrangements by SBH, corresponded to one patient with Hodgkin’s disease and two patients with PTLD. Again, all of the patients with a germline pattern by SBH were polyclonal when analyzed by PCR. Table 4 summarizes the characteristics of the false positive patients. All of the patients had a diagnosis of lymphoma or ALP, so the PCR results may not indicate true false positives. In fact, patients 1, 2, and 4 were NHL without IgH rearrangements by SBH, yet all were clonal B cell proliferations by immunohistochemistry. The ALP patient with clonal rearrangement by PCR (Patient 3) showed clusters of immunoblasts in the biopsy raising the concern of a lymphoma. Finally, Patient 5 was a composite lymphoma consisting of Hodgkin’s disease and peripheral T cell lymphoma. All five patients either had or were likely to have a monoclonal B cell population. The real false positive rate, in fact, may be much lower than 8%. Table 5 shows a comparison between the seminested and single-step PCR procedures. The overall concordance between the two procedures was 87.1%. Approximately 84% of the monoclonal samples by the seminested procedure were monoclonal when using the single-step PCR analysis. Five of the monoclonal patients identified by the semi-nested PCR method were included Detection of IgH Rearrangements by PCR/Lozano et al. 1353 TABLE 4 Description of the False Positive Cases Case No. and Diagnosis SBH Semi-nested PCR Comments Clonal Clonal Clonal Clonal Clonal B-cell NHL by IHC B-cell NHL by IHC Morphologically suspicious for NHL B-cell NHL by IHC HD and T-cell lymphoma ~~ Germline Germline Germline TCR, TCRB 1. NHL-SL 2. NHL-MZL 3. ALP 4. NHL-DM 5. CL (HD/NHL) SL: small lymphocyfic;MZL: mantle zone lymphoma; DM. diffuse mixed; CL: composite lymphoma; IHC: immunohistochemistty; TCR,,: clonal TCR,, rearrangment. TABLE 5 Comparison of Semi-Nested and Single-Step PCR using a DGGE system, 6 were found to be monoclonal, and 7 non clonal (oligoclonal or polyclonal) (Fig. 2). Single-step PCR ~ Semi-Nested PCR Clonal Non-clonal Total 31 31 Clonal Non-clonal 26 (83.8%) 3 (9.7%) 5 (16.2%) 28 (90.3%) = 62; ovrrall concordance: = 87.1%. TABLE 6 Comparison Between Diffise and Follicular Lymphomas Clonal Type of MHL Southern hybridization PCR Percent* positive Diffuse Follicular 55 13 42 9 76.4 69.2 ‘Percent of clonal cases determined by Southern hybridization also shown to be clonal by PCR in the non clonal category by the single-step method. ‘Three of them were B cell lymphomas with IgH rearrangements by SBH, one was a PTLD with IgH rearrangements by SBH, and the other was an ALP with a germline pattern by SBH. In contrast, three polyclonal samples by seminested PCR were monoclonal using the single-step technique; two were B-NHLs with IgH rearrangements by SBH, and one was a patient with Hodgkin’s disease who was germline by SBH. Ninety percent of the polyclonal patients with the semi-nested procedure were also polyclonal using the single-step assay. Thirteen patients’ samples, analyzed by the seminested PCR method and PAGE, were considered inconclusive or difficult to interpret, mainly because there was a very faint band amid a polyclonal smear. Five of these cases were interpreted differently when analyzed by separate observers. When all of the 13 cases were reanalyzed DISCUSSION Alternatives to SBH for gene rearrangement analysis have been studied because of the time, labor, cost, and radioisotopes required for this technique. PCR based assays were found to be a realistic alternative to SBH for the assessment of clonal IgH gene rearrangements.”‘28””However, a comparison of the different PCR based studies reveals considerable variation in many aspects of experimental design and marked differences in the reported r e s ~ l t s . ~Therefore, -‘~ this study compared two different PCR assays for the detection of B cell clonality in a broad spectrum of lymphoid proliferations. The first approach was the use of a semi-nested procedure. Nested or semi-nested techniques allow the detection of template present in small amounts and increase the specificity of the r e a ~ t i o n . ~ ”In ~ ,addition, ” we analyzed some of the samples with a single-step procedure which was modified from a previously reported method.24 Assuming that the sensitivity and specificity of both techniques are similar, the single-step method, because of its simplicity and speed, would be a more desirable assay in screening for B cell clonality. The correlation between the two assays is very good in this study. Another comparative study using paraffin embedded tissue is currently underway. Of 35 patients evaluated, only one discrepancy was observed (overall concordance = 97%), suggesting that these two assays are also comparable for formalin fixed, paraffin embedded samples. One of the major problems with using PCR methods to analyzed IgH gene rearrangements is the number of false negative results, that is, rearrangements detected by SBH but missed by PCR. The reported frequency of false negativity ranges below 1F%9~’oand 35% to 45%.’2~’3*’7~’y In this series, the occurrence of false negative results was 25% when considering only NHL, and 28.5% when all of the clinical samples were included. Both percentages are in the published range for the use of semi-nested procedures and polyacrylamide gel analysis?~’“,22 There are sev- 1354 CANCER April 1,1996 / Volume 77 / Number 7 era1 explanations for false negative results. The VH-FR 111 should always be evaluated with morphologic, immunophenotypic, and clinical characteristics. consensus primer may not anneal well to all VH genes, A comparison of the reported studies shows including undescribed VH genes, and prevent efficient marked variation in the sensitivity of the PCR assays amplification.''~28~'9 It has been suggested that a panel of used, with results between 55% and more than VH family-specific primers would detect most VDJ rear~~~~9,10,12,13,21.U. , some of the latter results were from the rangements,8,23but the complexity of the assay is inPCR products in 1.5%or 2% agarose gels.9i21 analysis of creased and the large size of the template encompassed The lower percentage of detection in the NHL category by these primers makes them unsuitable for analyzing in our study, when compared with other investigations, degraded DNA extracted from fixed tissues. Segal and may be due, at least in part, to the study's strict intercolleagues recently reported a higher amplification effipretation of the PCR bands generated after electrophociency when using VH-FR 111 and JH consensus primers retic separation. We consider a case to be monoclonal than with VHFR I family-specific primers in low-grade, B when one or two distinct bands are identified after long cell lymphomas, excluding follicular lymphomas.21 polyacrylamide gel separation. Gels with less separatAnother aspect to consider is that PCR methods, uning power may give rise to false positive results. like SBH, can only detect complete VDJ rearrangements Since the consensus primers used in this study can which are in the correct orientation. A DJH rearrangeamplify the CDR 111 of most B cells present, the products ment, an unconventional rearrangement, or an interchroof a small clonal population may be obscured by the mosomal translocation involving the IgH locus may be of polyclonal products. Most of the interobserver smear detected by SBH but missed by PCR?," Another source variability in this study was due to faint bands in a polyof false negative results lies in the rearrangement process clonal smear. The use of different and more discriminatitself. Critical nucleotides can be deleted from the 3' end ing gel analysis systems has been proposed by other^.'^," of the VH gene or the 5' end of the JH gene, thus preThe use of DGGE has been helpful in the interpretation venting PCR amplifi~ation.~,~ This may explain why two of some cases in which a clear conclusion is impossible NHLs with demonstrable IgH rearrangements by SBH using even long polyacrylamide gels. DGGE separates were negative using the semi-nested PCR method, and DNA fragments based on the denaturing characteristics clonal using the single-step technique with a JH primer inherent to their sequence, so fragments which differ only farther from the recombination site than the JH-nested in a single nucleotide can be primer. Deletions that affect priming by the JH-nested Some studies have reported that the type of B cell primer may not interfere with the more 3' primer used neoplasm influences the ability to detect IgH gene rearin the single-step procedure. In addition, somatic mutarangements by PCR with currently-used consensus primtions in critical nucleotides may prevent the oligonucleoers.12.lfi,21.22 Follicular lymphomas may have the lowest tide primers from hybridizing with target detection level, presumably due to somatic mutations Conversely, the high sensitivity of the PCR assays which may affect primer annealing. In this series, 13 pamay have allowed the detection of clonality in some cases tients with follicular lymphoma showed IgH rearrangein which SBH was unable to detect rearrangements. It ment by SBH; 9 were monoclonal by PCR (69.2%) (Table should be noted, however, that the sensitivity of PCR us6).Thus, follicular lymphomas may indeed have a slightly ing consensus primers depends on the proportion of lower rate of clonality by PCR compared with diffuse polyclonal B cells present in the sample. In this study, lymphomas (76.4%), but the small number in this series the false positive rate was 8%, but these false positive does not allow a definite conclusion. cases all had a diagnosis of lymphoma or ALP, which may In summary, it was concluded that PCR analysis of not indicate true false positive results (Table 4). In fact, the IgH rearrangement is an efficient technique for the two NHLs (small lymphocytic and mantle cell NHLs) with initial determination of clonality in B-lymphoid proliferaB cell clonality by immunohistochemistry, and a case of tions, with 75% of SBH-positive NHLs also detectable by ALP that was suspicious for occult lymphoma, belong to this technique. The overall concordance between the this false positive category. The two cases with TCR-B semi-nested and single-step procedures was good, with rearrangements by SBH and IgH monoclonality by PCR the single-step method having the advantage of simplicity corresponded to a diffuse mixed small and large cell and speed. A negative result by either method may be an lymphoma of B cell phenotype and a case of composite indication of SBH. The DGGE system is useful for the lymphoma with a HD and a T cell lymphoma component. assessment of clonality in assays that are difficult to interThe reactive and miscellaneous categories showed 100% pret in PAGE, since DGGE enhances the detection of miconcordance in the 20 germline samples studied (Table nor clonal B cell populations and biallelic clonal rear3 ) . As clonality does not always equate m a l i g n a n c ~ , ~ ~rangements ,~~ which give rise to products of almost identical size. In some cases, a combination of different results of genotypic analysis, whether by SBH or PCR, Detection of IgH Rearrangements by PCR/Lozano et al. techniques can achieve a high level of sensitivity and specificity. 17. REFERENCES 1. 2. 3. 4. 5. ti. 7. 8. 9. 10 11 12. 13. 14. 15. 16. Early P, Huang H, Davis M, Calame K,Hood L. An immunoglobulin heavy chain variable region gene is generated from three segments of DNA: VH, D, and JH. Cell 1980;19:98192. Tonegawa S. Somatic generation of antibody diversity. Nature 1983;302:575-81. Walter MA, Surti U, Hofker MH, Cox DW. The physical organization of the human immunoglobulin heaw chain complex. EMBO J 1990;9:3303-13. Komori T, Okada A, Stewart V, A t FW. Lack of N regions in antigen variable region genes of TdT-deficient lymphocytes. Science 1993;26 1:1171-5. Cleary ML, Meeker TC, Levy S, Lee E, Trela M, W a r J, et al. Clustering of extensive somatic mutations in the variable region of an immunoglobulin heavy chain gene from a human B-cell lymphoma. Cell 1986;44:97- 106. Levy R, Levy S, Cleary ML, Carroll W, Kon S, Bird J, et al. Somatic mutations in human B-cell tumors. Inzmunol Rev 1987;96:43-58. Yamada M, Hudson S, Tournay 0, Bittenbender S, Shane SS, Lange B, et al. Detection of minimal disease in hematopoietic malignancies of the B-cell lineage by using thirdcomplementarity-determining region (CDR-111)-specific probes, ?roc Natl Acad Sci USA 1989;86:5123-7. Deane M, Norton JD. Detection of immunoglobulin gene rearrangement in B-cell neoplasias by polymerase chain reaction gene amplification. Leuk Lymphoma 1991;5:9-22. Wan JH, Trainor KJ, Brisco MJ, Morley AA. Monoclonality in B-cell lymphoma detected in paraffin wax embedded sections using the polymerase chain reaction. J CIin Path01 1990;43:888-90. Deane M, Norton JD. Immunoglobulin gene “fingerprinting’’: an approach to analysis of B lymphoid clonality in iymphoproliferative disorders. Br J Haematol 199I; 77274431. ‘Trainor KJ, Brisco MI, Wan JH, Neoh S, Grist S, Morley AA. Gene rearrangement in B- and T-lymphoproliferative disease detected by the polymerase chain reaction. Blood 1991;78:192-96. Ben Ezra J. Variable rate of detection of immunoglobulin heavy chain V-D-J rearrangement by PCR a systematic study of 41 B-cell non-Hodgkin’s lymphomas and leukemias. Leuk Lymphoma 1992;7:289-95. Liang R, Chan V, Chan TK, Wong T, Todd D. Detection of immunoglobulin gene rearrangements in B-cell lymphomas by polymerase chain reaction gene amplification. Hematol Oncol 1992;lO;149-54. Segal GH, Wittwer CT, Fishleder AJ, Stoler MH, Tubbs RR, Kjeldsberg CR. Identification of monoclonal B-cell populations by rapid cycle polymerase chain reaction. A practical screening method for the detection of immunoglobulin gene rearrangements. A m J Pathol 1992;141:1291-97. Davis TH, Yockey CE, Balk SP. Detection of clonal immunoglobulin gene rearrangements by polymerase chain reaction amplification and single-strand conformational polymorphism analysis. Am J Path01 1993;142:1841-47. Diss TC, Peng H, Wotherspoon AC, Isaacson PG, Langxing P. Detection of monoclonality in low-grade B-cell lymphomas 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 1355 using the polymerase chain reaction is dependent on primer selection and lymphoma type. J Pathol 1993;169:291-95. Inghirami G, Szabolcs MJ, Yee HT, Corradini P, Cesarman E, Knowles DM. detection of immunoglobulin gene rearrangement of B-cell non-Hodgkin’s lymphomas and leukemias in fresh, unfixed, and formalin-fixed, paraffin-embedded tissue bv- polvmerase chain reaction. Lab Invest - . 1993;681746-57. Reed TJ, Reid A, Wallberg K,O’LearyTJ, Frizzera G. Determination of B-cell clonality in paraffin-embedded lymph nodes using the polymerase chain reaction. Diag Mol Patkol 1993;2:42-49. Slack DN, McCarthy KP, Wiedemann LM, Sloane JP. Evaluation of sensitivity, specificity, and reproducibility of an optimized method for detecting clonal rearrangements of immunoglobulin and T-cell receptor genes in formalin-fixed, paraffin-embedded sections. Diag Mol Pathol 1993;2:223-32. Chen YT, Whitney KD, Chen Y. Clonality analysis of B-cell lymphoma in fresh-frozen and paraffin-embedded tissues: the effects of variable polymerase chain reaction parameters. Modern Pathol 1994;7:429-34. Segal GH, Jorgensen T, Masih AS, Braylan RC. Optimal primer selection for clonality assessment by polymerase chain reactiop analysis: 1. Low grade B-cell lymphoproliferative disorders of nonfollicular center cell type. Hum Pathol 1994;25:1269-75. Achille A, Scarpa A, Montresor M, Scardoni M, Zamboni G, Chilosi M, et al. Routine application of polymerase chain reaction in the diagnosis of monoclonality of B-Cell lymphoid proliferations. Diag Mol Pathol 1995;4:14-24. Lehman CM, Sarago C, Nasim S, Comerford J, Karcher DS, Garrett CT. Comparison of PCR with Southern hybridization for the routine detection of immunoglobulin heavy chain gene rearrangements. A m J Clin Pathol 1995; 103:171-76. Sioutos N, Bagg A, Michaud GY, Irving SG, Hartmann DP, Siragy H, et al. Polymerase chain reaction versus Southern blot hybridization. Detection of immunoglobulin heavychain gene rearrangements. Diag Mol Pathol 1995;4:8- 13. Sambrook J, Fritsch EF, Maniatis TM. Molecular cloning: a laboratory manual. 2nd ed. New York: Cold Spring Harbor Laboratory, 1989. Chou Q, Russell M, Birch DE, Raymond J, Bloch W. Prevention of pre-PCR miss-priming and primer dimerization improves low-copy-number amplifications. Nucleic Acid Res 1992;20:1717-23. Greiner TC, Raffeld M, Lutz C, Dick F, Jaffe ES. Analysis of T-cell receptor-7 gene rearrangements by denaturing gradient gel electrophoresis of GC-clamped polymerase chain reaction products. A m J Pathol 1995; 146:46-55. Chan WC, Greiner TC. Diagnosis of lymphomas by the polymerase chain reaction. Am J Clin Pathol 1994;102:273-74. Medeiros LJ, Weiss LM. The utility of the polymerase chain reaction as a screening method for the detection of antigen receptor gene rearrangements. Hum Pathol 1994;25:1261-63. Weiss LM, Wood GS, Trela MS, Warnke FZA, Sklar J. Clonal T-cell populations in lymphoid papulosis. Evidence for a lymphoproliferative etiology in a clinically benign disease. N Engl J Med 1986;315:475-79. Fishleder A, Tubbs R, Hesse B, Levine H. Uniform detection of immunoglobulin gene rearrangements in benign lymphoepithelial lesions. N Engl J Med 1987;316:1118-21. Tierens A, Lozano MD, Wickert RS, Chan WC, Greiner TC. High resolution analysis of immunoglobulin heavy chain gene rearrangements using denaturing gradient gel electrophoresis. Diag Mol Puthol 1996, in press.