OUP user menu

Improved Detection Rate of Cytogenetic Abnormalities in Chronic Lymphocytic Leukemia and Other Mature B-Cell Neoplasms With Use of CpG-Oligonucleotide DSP30 and Interleukin 2 Stimulation

Min Shi MD, PhD, Matthew J. Cipollini MA, Patricia A. Crowley-Bish, Anne W. Higgins PhD, Hongbo Yu MD, PhD, Patricia M. Miron PhD
DOI: http://dx.doi.org/10.1309/AJCP7G4VMYZJQVFI 662-669 First published online: 1 May 2013

Abstract

Detection of cytogenetic abnormalities requires successful culture of the clonal population to obtain metaphase chromosomes for study, and as such, has been hampered by low mitotic indices of mature B cells in culture. Our study presents data on the improved abnormality detection rate with the use of a CpG-oligonucleotide/interleukin 2 (OL/IL-2) culture protocol for mature B-cell neoplasms, including chronic lymphocytic leukemia (CLL) and non-CLL specimens. The increased detection rate of abnormalities, compared with unstimulated culture and traditional pokeweed mitogen culture, was statistically significant for both CLL and non-CLL neoplasms. For CLL specimens, our data also showed that for cytogenetically visible aberrations, OL/IL-2 was as, if not more, sensitive than detection with interphase fluorescence in situ hybridization (iFISH). Use of OL/IL-2 allowed a number of abnormalities to be detected, which were not covered by specific iFISH panels, especially balanced translocations. Therefore, OL/IL-2 stimulation improves diagnostic sensitivity and increases discovery rate of novel prognostic findings.

Key Words:
  • Cytogenetics
  • B-cell lymphoma
  • CpG-oligonucleotide DSP30

Mature B-cell neoplasms are common hematologic malignancies in the western hemisphere, with chronic lymphocytic leukemia (CLL) being the most prevalent.1 Cytogenetic analysis is a useful adjunct to morphologic assessment for both diagnosis and prognosis, as has been particularly well-established in CLL.25 However, conventional cytogenetic analysis has been hampered by a low mitotic index of mature B cells in culture.

For specific established recurring cytogenetic abnormalities, interphase fluorescence in situ hybridization (iFISH) assays are used to detect abnormalities in cells that do not divide readily in culture. FISH probes for a number of translocations diagnostic in B-cell neoplasms are commercially available [eg, t(11;14)(q13;q32) in mantle cell lymphoma, t(14;18)(q32;q21) in follicular lymphoma] as are CLL-specific FISH panels. In CLL, iFISH with a 4-assay panel (for detection of trisomy 12, and deletion of 13q, ATM, and TP53) has an approximately 80% abnormality detection rate and provides useful prognostic information because TP53 and ATM deletions correlate with poor outcome whereas isolated 13q deletion correlates with a favorable outcome.2 However, iFISH is limited because of its targeted nature and thus cannot provide the comprehensive genomic assessment afforded by metaphase cytogenetic analysis of banded chromosomes. In addition, iFISH cannot contribute to discovery of new prognostic abnormalities that can be revealed by a complete cytogenetic analysis of metaphase chromosomes.

B-cell mitogens have been used to stimulate division of cells for chromosome analysis. Clinical cytogenetic laboratories have commonly used pokeweed mitogen (PWM), 12-O-tetradecanoyl-phorbol-13-acetate, and lipopolysaccharide with some success. For CLL, these mitogens typically yield an abnormality detection rate of approximately 40% to 50%.6 Recently, clinical laboratories have started to assess protocols using synthetic CpG-oligodeoxynucleotides (CpG-ODN) to stimulate cell division in CLL specimens and enhance the detection rate of clonal aberrations710 after research showed that these oligonucleotides induce proliferation, cytokine production, and high-affinity interleukin 2 (IL-2) receptor expression in CLL cells11,12 through interaction with Toll-like receptor 9 on B cells.1315

In the United States, studies largely from 1 group8,9 showed improved abnormality detection with a protocol involving CpG-ODN in combination with PWM and phorbol myristate acetate. Because prior research studies showed that IL-2 prevents cell death of CLL cells activated by CpG-ODN, European groups introduced a protocol that combines IL-2 with CpG-ODN and have shown success with this culture method in CLL.16

Mature B-cell neoplasms are heterogeneous, and in addition to CLL, include follicular lymphoma, mantle cell lymphoma, marginal zone lymphoma, lymphoplasmacytic lymphoma, and hairy cell leukemia. Similar to CLL cells, the tumor cells in this group of disease generally do not respond well to the standard B-cell mitogens. CpG-oligonucleotide stimulation is not specific to CLL cells and has been shown to stimulate division of B cells more generally.11,12,1719 As such, we also investigated the impact of the combined use of CpG-oligonucleotide and IL-2 (OL/IL-2) protocol for detection of cytogenetic abnormalities in other mature B-cell neoplasms.

Herein we present data from our clinical cytogenetics laboratory that support improved abnormality detection for both CLL and other mature B-cell neoplasms with the use of OL/IL-2. Our abnormality detection rate for OL/IL-2 cultures is significantly higher than that for both PWM cultures and unstimulated (overnight Colcemid [ONC]) cultures. In addition to increasing the overall abnormality detection, our data also show that OL/IL-2 application detects more subclones with higher complexity than the other methods, with implications for possible additional prognostic stratification. Notably, many of the abnormalities detected with OL/IL-2 would not be detected with current standard FISH panels.

Materials and Methods

Patients

Peripheral blood or bone marrow samples from 81 patients with CLL and 73 patients with another mature B-cell neoplasm were processed for metaphase chromosome analysis and iFISH. iFISH was performed on only a subset of these patients, based on request or metaphase chromosome findings. Specimens were obtained over a 2-year period, and included all cases of confirmed disease based on morphologic and/or flow cytometric findings. All specimens represent a unique patient.

Metaphase Chromosome Analysis

Cultures were performed on cells from peripheral blood or bone marrow using MarrowMax complete medium (Invitrogen, Carlsbad, CA) at a concentration of 106 cells/mL in parallel under 3 culture conditions: overnight dilute Colcemid (Invitrogen), 0.01 μg/mL with no mitogen; PWM (Sigma, Saint Louis, MO), 0.1 mg/mL; and CpG-oligonucleotide DSP30 (TIB MolBiol, Adelphia, NJ), 2 μmol/L plus IL-2 (Sigma), 0.04 μg/mL.

Metaphase chromosomes were stained using a Giemsa and trypsin solution (GTG-banding). Ten metaphase cells from each of the 3 cultures (ONC, PWM, and OL/IL-2) were analyzed. A clone was defined as 2 cells with the same additional or structurally rearranged chromosome or as 3 cells with the same chromosome loss.

FISH

Interphase FISH was performed on uncultured nuclei. The CLL FISH panel contained ATM (11q22.3), D12Z3 (12 cen), D13S319 (13q14.3), and TP53 (17p13.1). Other FISH probes used for detection of specific disease and/or confirmation of anomalies included a dual-color/dual-fusion BCL2/IGH@ (18q21/14q32) probe set, a dual-color/dual-fusion CCND1/IGH@ (11q13/14q32) probe set, a dual-color BCL6 (3q27) break-apart probe, a dual-color/dual-fusion MYC/IGH@ (8q24/14q32) probe set, and a dual-color IGH@ (14q32) break-apart probe. All probes were obtained from Vysis (Abbott Molecular, Abbott Park, IL). A minimum of 100 nuclei were studied for all assays.

The iFISH assay laboratory cutoffs are set at 3 standard deviations above the mean, as determined with an initial database of 10 normal controls. In the present series, the cutoffs are typically identical to, and sometimes more conservative than, the 99% confidence limit calculated for a binomial distribution. Specific cutoffs are as follows: 6% for ATM deletion, 8% for D13S319 deletion, 14% for TP53 deletion, 3% for trisomy 12, and 1% for dual-fusion assays.

Results

Cultures were performed in parallel on all specimens (81 CLL cases and 73 other mature B-cell neoplasm cases) in this study under the following 3 conditions: 24-hour unstimulated (ONC), 72-hour with PWM, and 72-hour with OL/IL-2. All cases had confirmed disease, but included new diagnoses as well as treated patients with minimal residual disease. Comparing abnormality rate based on study of cultured metaphase cells, the increase in detection rate for OL/IL-2 over the other 2 culture modalities was statistically significant (P < .05); at least 1 abnormal clone was detected in 84 (55%) of 154 OL/IL-2 specimens, 58 (38%) of 154 PWM specimens, and 18 (12%) of 154 ONC specimens Table 1. Thirty-six (23%) of 154 cases (17 CLL and 19 non-CLL cases) had unique detection of abnormalities with OL/IL-2.

View this table:
Table 1

CLL

Specimens from 81 CLL cases (53 bone marrow and 28 peripheral blood specimens) were studied. Interphase FISH with a CLL panel (Vysis/Abbott Molecular Laboratories) assessing trisomy 12 and deletion of 13q14.3, ATM, and TP53 was performed on 72 of the 81 CLL cases.

In 43 (53%) of 81 cases an abnormal clone was detected in at least 1 culture specimen by means of metaphase chromosome analysis. An abnormality was detected in 12 additional cases using iFISH with the CLL panel, for an overall abnormality rate of 57 (70%) of 81 cases. For cultured metaphase cells, the increase in detection rate for OL/IL-2 over the other 2 culture modalities was statistically significant (P < .05) with detection of at least 1 abnormal clone in 43 (53%) of 81 OL/IL-2 cultures, 30 (37%) of 81 PWM cultures, and 10 (12%) of 81 ONC cultures Table 2. (Note, 20 ONC cases had no growth; 15 of these cases were peripheral blood specimens.)

View this table:
Table 2

All the abnormal clonal abnormalities in PWM and ONC cultures were detected in the OL/IL-2 cultures. In contrast, a number of clonal abnormalities observed in OL/IL-2 cultures were not detected in PWM or ONC cultures. Eleven cases (14%) were abnormal only on OL/IL-2, and in 6 additional cases (7%), subclonal abnormalities were detected only in OL/IL-2 culture. Thus, for assessment of metaphase cells, in 17 cases (21%), specific abnormalities were detected only with OL/IL-2. Table 3 shows the results of cases with abnormalities detected only after OL/IL-2 culture.

View this table:
Table 3

The most common chromosome abnormality uniquely detected with OL/IL-2 was del(13)(q12q14), seen in 7 of the 17 cases. Trisomy 12 was seen twice in this cohort, monosomy 17 once, and deletion of the chromosome 17 short arm (17p–) once. These 11 abnormalities were confirmed on FISH with the Vysis CLL panel. Two additional structural abnormalities led to imbalances supported by the CLL FISH panel. One case had a (17;18) rearrangement resulting in deletion of the chromosome 17 short arm [45 chromosomes with −17,der(17;18)(q11.2;p11.2)]; the other case had a (13;21) translocation resulting in 13q14.3 deletion [t(13;21) (q14;q21)]. A third abnormality, t(1;14)(q32;q32) was supported by IGH@ chromosome-rearrangement FISH.

Recent data suggest that chromosomal translocations may be associated with poor prognosis in patients with CLL.20 Sixteen translocations were detected in 15 cases and 2 inversions in 2 cases, totaling 21% of the entire cohort and 38% of the abnormal cohort. Five of these rearrangements were detected solely in the OL/IL-2 culture Table 4. Of the translocations, 12 (75%) appeared to be balanced and 4 (25%) unbalanced. Only 1 translocation was recurring in our series, a whole arm translocation between the long arms of chromosomes 17 and 18. Both inversions appeared balanced.

In addition to having the highest abnormality detection rate among the 3 culture methods, OL/IL-2 also yielded the highest overall percentage of abnormal cells Figure 1. For the majority of abnormal OL/IL-2 CLL cases (30/43 cases, 70%), the abnormal clone comprised at least 60% of metaphase cells; in only 2 (5%) of 43 abnormal cases did the clone comprise less than 30% of the metaphase cells. In contrast, for the majority of CLL cases with an abnormal PWM-detected clone (12/30 cases, 40%), the abnormal clone comprised less than 30% of metaphase cells; in only 2 (7%) of 30 cases of abnormal PWM cases did the abnormal clone comprise 60% or more metaphase cells. For the vast majority of abnormal ONC cases (6/9 cases, 67%), the abnormal cells comprised less than 30% of the total metaphase cells.

Interphase FISH with the CLL panel had a higher overall abnormality detection rate (45/72 cases, 63%) than analysis of metaphase chromosomes (Table 2). However, deletion of 13q14.3, which is often cytogenetically cryptic, accounted entirely for the increased detection rate of iFISH. Of the 33 cases with 13q14.3 deletion on FISH, 14 had visible 13q deletions on metaphase chromosome analysis. All trisomy 12 (n = 14), ATM (n = 4), and TP53 (n = 3) deletion clones identified on iFISH were also identified on metaphase chromosome analysis in OL/IL-2 cultures. Thus, in this data set, iFISH did not increase the detection rate of cytogenetically visible aberrations compared with OL/IL-2.

Consistent with this finding, the percentage of abnormal metaphase cells obtained from OL/IL-2 cultures was higher than the percentage of abnormal uncultured nuclei seen on iFISH, reflecting positive selection in culture Figure 2. For 31 cases with an abnormality detected by assessment of both OL/IL-2 culture and iFISH, the average percentage of abnormal cells in OL/IL-2 culture was 68% whereas the average percentage of abnormal uncultured nuclei found on iFISH was 41% (P < .05). This supports comparable, if not increased, sensitivity of OL/IL-2 culture as compared to iFISH for detecting cytogenetically visible aberrations.

View this table:
Table 4
Figure 1

Distribution of abnormal metaphase cells by culture method in chronic lymphocytic leukemia (CLL) cases. Black bars represent the percentage of abnormal metaphase cells from those studied after culture with CpG-oligonucleotide DSP30 plus interleukin 2 stimulation (OL/IL-2). White bars represent the percentage of abnormal metaphase cells from those studied after culture with pokeweed mitogen stimulation (PWM). Gray bars represent the percentage of abnormal metaphase cells from those studied after overnight culture with overnight dilute Colcemid (ONC) and no mitogen.

The disease range over which OL/IL-2 culture detected an abnormality was also similar to iFISH. For both methods, the minimum tumor burden (determined with flow cytometry) associated with an abnormality was 0.8%. For cases with greater than 5% tumor burden, the abnormality detection rate was 57% for OL/IL-2 and 63% for iFISH; for cases with less than 5% tumor burden, the detection rate was 25% for both OL/IL-2 and iFISH.

Assessment of metaphase chromosomes has the additional advantage over iFISH of allowing detection of abnormalities not covered by specified FISH panels. In this series, 24 cases (29% of total cohort, 42% of abnormal cohort) had abnormalities in the OL/IL-2 culture that were not detectable with the more extensive CLL FISH panel that includes testing for 6q deletion. These included chromosomal translocations, inversions, deletions, additions (unidentified additional material), loss of X (n = 2), and trisomy 7 (n = 1).

Interphase FISH also underestimates overall complexity of the genome, known to play a key role in prognosis in CLL.6,2123 With OL/IL-2, 9 cases (11% of total cohort, 16% of abnormal cohort) had 3 or more aberrations; 5 of these cases showed 3 aberrations, 1 showed 4 aberrations, and 3 showed 5 or more aberrations. In contrast, only 1 case showed more than 2 aberrations on iFISH analysis.

Other Mature B-Cell Neoplasms

Specimens from 73 other mature B-cell neoplasms (62 bone marrow and 11 peripheral blood specimens) were studied, which included 21 mantle cell lymphoma, 11 follicular lymphoma, 12 lymphoplasmacytic lymphoma, 8 extranodal marginal zone lymphoma, 6 splenic marginal zone lymphoma, 2 hairy cell leukemia, and 13 unclassified mature B-cell lymphoma cases Table 5.

An abnormal clone was detected in 42 cases (58%). The increased detection rate for OL/IL-2 was statistically significant (P < .05), with at least 1 abnormal clone detected in 41 (56%) of 73 OL/IL-2 cultures, compared with 28 (38%) of 73 PWM cultures, and 8 (11%) of 73 ONC cultures (Table 5). Only 1 abnormal case was not detected on OL/IL-2; technologist error is suspected, because in this case, the abnormal clone was detected in both the ONC and the PWM cultures [mantle cell lymphoma with t(11;14)(q13;q32)], but yielded no cells on the OL/IL-2 culture. Although the size of this study does not allow meaningful statistical analyses by specific disease type, the OL/IL-2 mitogen outperformed PWM and ONC for each disease category, except for the 2 cases of hairy cell leukemia that had no detectable anomaly with any method (Table 5).

Figure 2

Distribution of abnormal cells detected on culture with CpG-oligonucleotide DSP30 plus interleukin 2 (OL/IL-2) and interphase fluorescence in situ hybridization (iFISH) in chronic lymphocytic leukemia (CLL) cases. Black bars represent the percentage of abnormal metaphase cells from those studied after OL/IL-2 culture. White bars represent the percentage of abnormal nuclei from those counted after iFISH of uncultured specimen.

In addition to yielding the highest number of cases in which at least 1 abnormal clone was detected, OL/IL-2 mitogen cultures also yielded the highest number of clonal aberrations. In 5 cases, in addition to the 14 in which OL/IL-2 was the only culture method to yield any abnormalities, 1 or more subclonal aberrations were detected solely in OL/IL-2. Thus, among the 42 cases that were abnormal using at least 1 culture method, 1 or more abnormalities were detected solely in OL/IL2 cultures in 19 (45% of abnormal) cases.

View this table:
Table 5
Figure 3

Distribution of abnormal metaphase cells by culture method in mature B-cell neoplasms other than chronic lymphocytic leukemia (CLL). Black bars represent the percentage of abnormal metaphase cells from those studied after culture with CpG-oligonucleotide DSP30 plus interleukin 2 (OL/IL-2) stimulation. White bars represent the percentage of abnormal metaphase cells from those studied after culture with pokeweed mitogen (PWM) stimulation. Gray bars represent the percentage of abnormal metaphase cells from those studied after overnight culture with dilute Colcemid (ONC) and no mitogen.

Among the 14 cases in which OL/IL-2 was the only culture medium to yield an abnormal clone were 4 specimens with t(11;14)(q13;q32), 3 with t(14;18)(q32;q21), 2 with i(6) (p10), one with trisomy 12, 1 with i(17)(q10), and 1 with trisomy 21, all of which are recognized recurrent cytogenetic aberrations and consistent with the pathologic diagnoses. Two of the 14 specimens had less specific abnormalities: 1 had a dup(11)(q13q23) abnormality and the other had a der(6)t(3;6) (q13;p12) abnormality. For the 5 additional cases with subclonal findings seen only in the OL/IL-2 culture, the subclonal chromosome aberrations were not recognized recurrent aberrations but did serve to identify clonal progression.

In 3 (7% of abnormal) cases, an abnormality was detected in the PWM culture that was not detected in the OL/IL-2 culture. One case was the no-growth OL/IL-2 culture that was suspected to be technologist error. In the second case, a primary follicular lymphoma abnormality, t(14;18)(q32;q21), was detected in both PWM and OL/IL-2 cultures, but presumed secondary abnormalities, add(X)(p11.2) and del(10) (q24), were uniquely detected in 3 PWM cells. The third case had 2 independent clones detected in PWM, but only 1 of the 2 clones was detected in the OL/IL2 culture.

As found in CLL, OL/IL-2 also yielded the highest overall percentage of abnormal cells Figure 3. For the majority of abnormal OL/IL-2 cases (17/41 cases, 41%), the abnormal clone comprised 60% or more metaphase cells; in only 9 (22%) of 41 abnormal cases did the clone comprise less than 30% of the metaphase cells. In contrast, for the majority of abnormal PWM cases (13/28 cases, 46%), the abnormal clone comprised less than 30% of metaphase cells; in only 7 (25%) of 28 cases did the abnormal clone comprise 60% or more metaphase cells. For the majority of abnormal ONC cases (6/8 cases, 75%), the abnormal cells comprised less than 30% of the total metaphase cells.

Specimen culture failure rate was highest for ONC cultures and lowest for PWM, but not statistically different than OL/IL-2. For ONC cultures, 9 (12%) had no growth; all were peripheral blood specimens. Three OL/IL-2 cultures (4%) had no growth; 1 was a peripheral blood and 2 were bone marrow specimens, 1 of which most likely represents technologist error. Only 1 peripheral blood specimen had no growth in PWM culture.

Interphase FISH was performed on 19 cases in a case-specific fashion (clinician request, clarification of chromosome aberration, suspicion of disease-specific aberration). For 9 cases that were abnormal on chromosome analysis, the iFISH was consistent with the metaphase chromosome result and did not provide additional information. For 10 cases with normal metaphase chromosome results, iFISH detected an abnormality in 7 cases: CCND1-IGH@ in 4 of 4 mantle cell lymphomas and BCL2-IGH@ in 3 of 4 follicular lymphoma cases. (MALT1-IGH@ FISH was performed at pathologist’s request in 2 cases of low-grade B-cell lymphoma with marginal zone lymphoma in the differential—neither had evidence of rearrangement.) These 7 cases had less than 5% involvement of nucleated cells in bone marrow on pathologic analysis and abnormalities detected with iFISH ranged from 2% to 10%. Thus, iFISH (with dual-fusion assays) was more sensitive than the metaphase chromosome analysis in mantle cell lymphoma and follicular lymphoma, and should still be performed in the context of low-level disease with a normal metaphase cytogenetic result.

Discussion

Our data show improved abnormality detection in both CLL and other mature B-cell neoplasms with the use of OL/IL-2 mitogen, supporting previously published data that showed improved abnormality detection in CLL with the use of a related CpG protocol,8,9 and also showed the benefit of this mitogen for other mature B-cell neoplasms. For both CLL and other mature B-cell neoplasms, the overall abnormality detection rate with the OL/IL-2 method increased to 55% compared with 37% with PWM and 12% with no mitogen. Use of OL/IL-2 also increased the detection of subclones in both specimen groups. In CLL, 40% of the abnormal cases (21% of total cases) had unique detection of an abnormality; for the non-CLL cases, 45% of abnormal cases (26% of total cases) had unique detection. In keeping with these findings, our data also show that in cases in which the abnormality was detected using multiple culture protocols, the percentage of abnormal cells was always highest with the OL/IL-2 protocol.

For CLL cases, no lymphoid clonal abnormality was missed by the study of OL/IL-2 culture compared with detection with PWM or ONC. For the other mature B-cell neoplasms, OL/IL-2 missed 2 lymphoid clones that were detected on PWM, 1 because of culture failure and the second in a case in which 2 independent clones were detected on PWM, only 1 of which was detected on OL/IL-2. The clone missed on OL/IL-2 had +12, a known recurrent lymphoid abnormality. Also, in a case excluded from this dataset because of concomitant myeloid disease, both a myelodysplastic syndrome and a lymphoid clone were detected on PWM culture, but only the myeloid clone was detected on the OL/IL-2 culture. As such, PWM might still be considered as an additional culture to OL/IL-2 if the specimen is adequate enough to permit multiple culture methods, but it does not need to be for CLL cases.

Interphase FISH is a useful adjunct method for mature B-cell neoplasms, and in many settings, is offered as a more sensitive test. The selective advantage of the CLL cells in OL/IL-2 culture leads to increased sensitivity for detection of cytogenetically visible lesions compared with the sensitivity of the iFISH technique for uncultured nuclei. This result has potential implications for optimizing not only FISH assays, but also genomic microarray assays. However, for low tumor burden, short-term culture still may not be sufficient to allow the detection of clonal abnormalities on microarray analysis. For the non-CLL cases, iFISH of uncultured nuclei was more sensitive than chromosome analysis in 7 cases with less than 5% involvement of disease in the bone marrow.

Chromosome analysis remains an informative method in this era of FISH and genomic microarray testing. A number of the abnormalities in our series appeared to be balanced translocations and, as such, would not be detectable on genomic microarray analysis. Metaphase chromosome analysis is superior to FISH for overall genomic analysis, allowing the assessment of overall genomic complexity and the detection of abnormalities not targeted by specific FISH panels. In this series, 42% of the abnormal CLL specimens had an abnormality not detectable with the standard CLL FISH panel. Use of this superior B-cell mitogen should improve diagnostic testing and afford an opportunity for discovery of new prognostic findings.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
View Abstract