- Email: [email protected]
small lymphocytic lymphoma: a clinicopathologic analysis
Annals of Diagnostic Pathology 25 (2016) 15–19
Contents lists available at ScienceDirect
Annals of Diagnostic Pathology
Proliferation centers in bone marrows involved by chronic lymphocytic leukemia/small lymphocytic lymphoma: a clinicopathologic analysis☆ Jason C. Chang, MD, Alexandra M. Harrington, MD, Horatiu Olteanu, MD, PhD., Peter VanTuinen, PhD, Steven H. Kroft, MD ⁎ Department of Pathology, Medical College of Wisconsin, Milwaukee, WI
a r t i c l e
i n f o
Keywords: Chronic lymphocytic leukemia Proliferation center Bone marrow pattern
a b s t r a c t Objectives: Proliferation centers (PCs) are a characteristic finding in chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) lymph nodes, and their presence and extent in this site are not currently felt to be related to clinical course. In contrast, detailed clinicopathologic analyses of bone marrow (BM) PCs have not been previously reported. Methods: The PCs in 88 CLL/SLL BMs from 45 patients (pts) were graded (0-4) and were correlated with other morphologic, immunophenotypic, cytogenetic, and laboratory features. Results: Proliferation centers were present in 69 BMs (78%) from 32 pts (71%) and were distinct/prominent (grades 2-4) in 21 pts (47%), with the latter more commonly found in follow-up BMs (1/7 diagnostic BMs vs 49/81 follow-up BMs; P = .04). When present, PCs were most commonly graded as distinct nodules easily visible on ×10. No relationships were identified between PCs and any complete blood count parameter, serum lactate dehydrogenase or IgG levels, degree or pattern of BM involvement, blood morphology, CD38 and FMC7 expression by flow cytometry, or fluorescence in situ hybridization results, when the first encountered BM was considered for each patient. Conclusions: This represents the first detailed analysis of PCs in CLL/SLL BMs. In our tertiary center, PCs were seen frequently, in approximately three-fourths of cases. There were no statistical associations identified between PCs and cytogenetic, immunophenotypic, or other laboratory and morphologic findings. © 2016 Elsevier Inc. All rights reserved.
1. Introduction Chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/ SLL) is a clonal lymphoproliferative disorder composed of small, mature B cells in the peripheral blood (PB), bone marrow (BM), and lymph nodes. It is the most common leukemia in Western countries, representing approximately 30% of all leukemias [1]. Lymph nodes involved by CLL/SLL typically show effacement of architecture, with a diffuse or vaguely nodular pattern, and with pale, irregular nodules distributed in a darker cellular background [2]. The pale areas represent proliferation centers (PCs), which are concentrations of prolymphocytes and paraimmunoblasts, a spectrum of larger cells with more dispersed chromatin, conspicuous nucleoli, and more abundant cytoplasm [2]. Proliferation centers are thought to represent the site of Tcell–dependent and stromal cell–dependent immune responses, resulting
☆ Conflicts of interest and source of funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. ⁎ Corresponding author at: Department of Pathology, Medical College of Wisconsin, 9200 W. Wisconsin Ave, Milwaukee, WI 53226. Tel.: +1 414 805 8459; fax: +1 414 805 8735. E-mail address: [email protected] (S.H. Kroft). http://dx.doi.org/10.1016/j.anndiagpath.2016.07.011 1092-9134/© 2016 Elsevier Inc. All rights reserved.
in the selection and proliferation of clonal B cells [3]. Not surprisingly, PCs often demonstrate genetic and immunophenotypic differences compared with the surrounding small lymphocytes, including increased expression of Ki-67, CD71, MUM1/IRF4, HLA-DC, HLA-DR, CD20, and CD23, and down-regulation of IgD and CD9 [4-8]. Although some studies have not found a correlation between clinical features and the extent of PCs in lymph nodes [9], other studies suggest that the presence of confluent PCs in CLL/SLL or SLL lymph nodes was associated with a worse clinical course and/or specific cytogenetic abnormalities [2,10-11]. Bone marrow involvement by CLL/SLL may show nodular, interstitial, diffuse, or mixed patterns of infiltration [12-14]. The pattern of involvement in CLL/SLL was once regarded as an important prognostic factor, but more recent literature has described other, more powerful predictive markers, such as cytogenetics and molecular abnormalities. Although the presence of PCs, in the setting of the appropriate immunophenotype, is essentially diagnostic of CLL/SLL in the BM, it is generally considered to be uncommon in BM biopsy specimens. Despite the robust literature devoted to CLL/SLL infiltration patterns in BM and their relationship to prognosis, there are few data on the frequency of PCs in the BM; therefore, we sought to systematically study this morphologic finding in a large cohort of CLL/SLL BM biopsies. To the best of our knowledge, no previous studies have examined the presence
16
J.C. Chang et al. / Annals of Diagnostic Pathology 25 (2016) 15–19
and extent of BM PCs in correlation with cytogenetic abnormalities, immunophenotypic features, and other laboratory findings. 2. Materials and methods 2.1. Patients All patients with CLL/SLL diagnosed according to the World Health Organization 2008 criteria [15,16], who had a BM biopsy performed at Froedtert Hospital/Medical College of Wisconsin from 2006 to 2010, were reviewed. Cases with less than 5% BM involvement by CLL/SLL were excluded. Patient demographics, the complete blood count, serum lactate dehydrogenase (LDH), and IgG immunoglobulin were recorded. This study was approved by the institutional review board at the Medical College of Wisconsin. 2.2. Morphologic studies The hematoxylin and eosin–stained histologic sections of the BM biopsy and clot section were reviewed. The BM PCs were graded together by 2 of the authors (JCC and SHK) as follows: 0, absent; 1, present, but small and ill-defined, not visible on ×10 objective; 2, distinct, easily recognized on ×10 objective; 3, extensive; and 4, diffuse increase in prolymphocytes without discrete PC formation. Absent/ill-defined PCs equated to grades 0 to 1, and distinct/prominent PCs equated to grades 2 to 4. In addition, the patterns of BM infiltrate were classified into nodular, diffuse, interstitial, and mixed by previously described definitions [12-14]. In cases where more than one pattern existed, the primary and secondary patterns were noted. The degree of BM involvement by the CLL/SLL infiltrate was semiquantitatively classified into 4 categories: less than 25%, 26% to 50%, 51% to 75%, and more than 75%. If present, the location of the PCs was noted (ie, perisinusoidal, perivascular). The PB lymphocyte morphology was assessed by reviewing the corresponding Wright-stained PB smears. Cases were considered to have increased prolymphocytes if they comprised more than 10% of the lymphocytes. Cases were categorized as atypical morphology if more than 10% of lymphocytes show cleaved or irregular nuclear contours. Correlations with concomitant lymph node biopsies could not be performed as too few cases had such concomitant biopsies. 2.3. Flow cytometric analysis Flow cytometric studies were performed using previously described methods [17]. Briefly, specimens were analyzed by 4-color flow cytometry on a FACSCanto flow cytometer with FACSDiva software (Becton Dickinson, San Jose, CA). Antibodies relevant to the current analysis included CD38 and FMC7 (Becton Dickinson, Franklin Lakes, NJ). Isotype controls were performed in all cases. Cluster analysis was performed with Paint-A-Gate software (Verity House, Topsham, ME). CD38 and FMC7 expression was assessed by the degree of overlap between the neoplastic cluster and the same population in the isotype control tube, using a 20% cutoff at a 2% isotypic control threshold. 2.4. Fluorescence in situ hybridization A CLL/SLL fluorescence in situ hybridization (FISH) panel, consisting of a commercially available set of 5 probes (Vysis, Des Plaines, IL) to evaluate for trisomy 12, IgH rearrangements, and deletions 11q22, 13q14, and 17p13, was performed according to manufacturer's instructions. Two hundred interphase nuclei were analyzed for each probe. 2.5. Statistical analysis The histopathologic characteristics of the BM PCs were correlated with other morphologic, immunophenotypic, FISH, and laboratory
findings. The associations between 2 categorical variables were examined using the Fisher exact test. Comparisons between the means of continuous variables were performed using the t test. When examining for trends across unique patients, findings from the first BM available for review were considered in the analysis. Statistical significance was set at P b .05. Statistical analyses were performed using SPSS for Macintosh version 11 (SPSS, Chicago, IL) software.
3. Results 3.1. Cases The study population consisted of 88 BM examinations from 45 patients. The patients included 37 men and 8 women ranging from 48 to 81 years (median age, 63 years). Seven cases were diagnostic BMs, whereas the remaining 81 were follow-up assessments.
3.2. Morphologic features Histologically, PCs were present in 69 (78%) of 88 cases and 32 (71%) of 45 patients in the first BMs from each available for review. The PCs in BMs were histologically identical to their counterparts in lymph nodes, appearing as pale staining, vaguely nodular collections of prolymphocytes, paraimmunoblasts, and small lymphocytes that were noticeable against a background of dark staining typical small, round CLL/SLL lymphocytes. Nineteen (22%) cases had grade 1 PCs (Fig. 1A and B). Distinct/prominent PCs (grades 2-4) were relatively common and were observed in 50 (57%) of 88 cases and 21 (47%) of 45 patients at first BM review. Grade 2 PCs were the most common pattern, observed in 38 (43%) of 88 cases, whereas grade 3 and grade 4 PCs were present in 7 (8%) and 5 (6%) of 88 cases, respectively (Fig. 1C-F). The PCs were observed adjacent to marrow sinusoids in 13 (15%) of 88 cases and in a perivascular distribution in 14 (16%) of 88 cases (Fig. 2). (See Fig. 2.) Distinct/prominent PCs were seen in 1 (14%) of 7 diagnostic BMs and 49 (60%) of 81 follow-up BMs (P = .04). However, PCs were not a temporally consistent feature in patients. Many patients with multiple follow-up biopsies had some biopsies showing distinct/prominent PCs and others with complete absence of PCs (data not shown). The most common pattern of BM involvement was the mixed pattern, seen in 58 (66%) of 88 cases, followed by interstitial (23/88; 26%), diffuse (5/88; 6%), and nodular patterns (2/88; 2%). Among cases with a mixed pattern, the most common primary pattern was the interstitial pattern (40/58 cases; 69%) and the most common secondary pattern was the nodular pattern (35/58 cases; 69%). Overall, interstitial, nodular, and diffuse patterns were observed in 81 (92%) of 88, 49 (56%) of 88, and 16 (18%) of 88 cases, respectively, either as the sole pattern or as the primary or secondary pattern in a mixed pattern. Nodular and diffuse patterns did not coexist in any given case; whenever there was a mixed pattern with either the nodular or diffuse pattern, the other component was always the interstitial pattern. Distinct/prominent PCs were seen in 33 (66%) of 50 cases containing a nodular pattern compared with those cases with absent/ill-defined PCs 16 (42%) of 38 (P = .03; Table 1), but this association disappeared when first available BMs were analyzed for each patient (P = .377). In contrast, a diffuse or interstitial pattern of infiltrate was not associated with the presence of distinct/prominent PCs in total cases (both P = 1.00; Table 1) or first available BMs (P = .729, .421). When the degree of BM involvement by CLL/SLL was semiquantitatively assessed, 11 (13%), 13 (15%), 28 (32%), and 36 (41%) of 88 of cases showed less than 25%, 26% to 50%, 51% to 75%, and more than 75% involvement, respectively. However, the degree of involvement did not correlate with the presence of distinct/prominent PCs (P = .62).
J.C. Chang et al. / Annals of Diagnostic Pathology 25 (2016) 15–19
17
Fig. 1. Morphologic classification of PCs in BM (hematoxylin and eosin). (A) Small/ill-defined PC (grade 1; ×50). (B) Small/ill-defined PC (grade 1; ×100). (C) Extensive PCs (grade 3; ×20). (D) Grade 3 PC (×50). (E) Diffuse increase in prolymphocytes without PC formation (grade 4; ×50). (F) Diffuse prolymphocytosis (grade 4; ×100).
3.3. Fluorescence in situ hybridization Complete FISH panels were performed on 43 patients on a total of 81 BMs; 1 additional patient had FISH performed only for del(11q) and trisomy [12]. Cytogenetic abnormalities were detected by FISH in 32 (71.1%) of 44 patients and 64 (79%) of 81 cases. IgH abnormalities and deletion 13q14 were the most common cytogenetic abnormalities, each present in 16 (37%) of 43 patients. The remaining cytogenetic abnormalities, in order of frequency, included trisomy 12 in 11 (26%), del 17p13 (p53) in 10 (23%), and deletion 11q22 (ATM) in 9 (21%) of 43 patients.
Fig. 2. Distribution of PCs in BM (hematoxylin and eosin). (A) Perisinusoidal PC (×20). (B) Perivascular PC (×50).
When considering all BMs with FISH (n = 81), distinct/prominent PCs were associated with an absence of trisomy 12 (Table 1), but this relationship disappeared when only analyzing the first BM of the patients (P = .175). The presence of distinct/prominent PCs was not significantly associated with other FISH abnormalities for total BMs (Table 1) or unique patients. 3.4. Immunophenotyping CD38 and FMC7 expressions were assessed in 47 and 69 cases by flow cytometry, respectively. This immunophenotypic data are presented
Fig. 3. The PB morphology examples (Wright Giemsa). (A) Case with increased PB prolymphocytes (×100). (B) Case with atypical PB cytology (×100).
18
J.C. Chang et al. / Annals of Diagnostic Pathology 25 (2016) 15–19
Table 1 Comparison of the BM pattern of CLL infiltration, cytogenetic and immunophenotypic findings, and mean clinical and laboratory parameters between BMs with absent/ill-defined PCs and BMs with prominent/distinct PCs
BM infiltration pattern Nodular pattern Diffuse pattern Interstitial pattern FISH findings del(11q) Trisomy 12 del(13q) del(17p) IgH Immunophenotype CD38+ FMC-7+ Clinical/laboratory parameters Age (y) Hemoglobin (g/dL) Platelet count (thousand/μL) Lymphocyte count (thousand/μL) γ-Globulin (g/dL) LDH (U/L)
Absent/ill-defined PCs (n = 38)
Distinct/prominent PCs (n = 50)
P
16/38 (42%) 7/38 (18%) 35/38 (92%)
33/50 (66%) 9/50 (18%) 46/50 (92%)
.03 1.00 1.00
8/36 (22%) 14/36 (39%) 10/36 (28%) 8/36 (22%) 16/36 (36%)
7/45 (16%) 7/45 (16%) 18/45 (40%) 14/45 (31%) 19/45 (42%)
.57 .02 .34 .46 1.00
12/20 (60%) 3/35 (9%)
22/27 (81%) 0/34 (0%)
.27 .24
62.2 11.3 127
62.0 11.1 90
.87 .75 .02
29.1
13.3
.08
0.65 238
0.55 224
.21 .46
in Table 1. No statistically significant relationships were identified between CD38 or FMC7 expression and PCs.
3.5. Clinical and laboratory data Of the clinical and laboratory data, only platelet count was significantly associated with the presence of distinct/prominent PCs in BMs. Patients whose BMs showed distinct/prominent PCs had a mean platelet count of 90 × 103/μL, compared with 127 × 10 3/μL in those patients whose BMs showed absent/ill-defined PCs (P = .02; Table 1); however, this relationship was not present when analyzing only the first available BM from each patient (P = .347). No other clinical and laboratory data were significantly correlated with the presence of distinct/prominent PCs in BMs (Table 1).
3.1. Peripheral blood findings Fifty-six cases had corresponding PB smears available for review. Fifteen (27%) of 56 cases showed more than 10% prolymphocytes in the PB (Fig. 3A). Increased PB prolymphocytes were present in 6 (20%) of 30 cases of BMs with distinct/prominent PCs and 9 (35%) of 26 BMs with absent/indistinct PCs (P = .24). Increased PB prolymphocytes were strongly associated with the presence of trisomy 12 (P b .01) but no other cytogenetic abnormalities (Table 2). Increased PB prolymphocytes were associated with more extensive degree of CLL BM infiltration and higher LDH values (Table 2). Thirteen (23%) of 56 cases showed atypical PB morphology (Fig. 3B). Atypical PB morphology was present in 10 (33%) of 30 cases of BMs with distinct/prominent PCs and 3 (12%) of 26 cases of BMs with absent/indistinct PCs (P = .06), but this trend was not evident when the analysis was restricted to first encountered BMs (P = .157). Atypical PB cytology was associated with deletion 13q14 (P = .01), although this association was also not evident on analysis of first encountered BMs (P = .173; Table 3). No statistically significant associations were found between atypical PB morphology and other cytogenetic abnormalities or clinical parameters (Table 3).
Table 2 Comparison of cytogenetic findings, degree of infiltration, and mean laboratory parameters between cases with and without increased PB prolymphocytes
del(11q) Trisomy 12 del(13q) del(17p) IgH translocation Degree of BM CLL infiltrate (%) LDH (U/L) γ-Globulin (g/dL)
Without increased PB prolymphocytes (n = 41)
Increased PB prolymphocytes (n = 15)
P
11/39 (28%) 5/38 (13%) 17/38 (45%) 12/38 (32%) 17/38 (45%) 63
2/15 (13%) 12/15 (80%) 2/14 (14%) 5/14 (36%) 6/14 (43%) 84
.31 b.01 .06 1.00 1.00 .001
210 0.60
312 0.62
b.001 .85
4. Discussion Although the clinical significance of PCs in CLL/SLL lymph node biopsies has been previously studied, no systematic studies to date have examined the frequency of PCs in BMs and the significance of this marrow finding. For that reason, this study aimed to assess the presence of PCs in the BM and correlate the extent of PCs to other morphologic, immunophenotypic, cytogenetic, and laboratory findings that are well established as prognostic indicators in CLL/SLL. The present study demonstrates that PCs are a relatively frequent feature in BMs involved by CLL/SLL. In our study, PCs were present in 78% of marrows overall and 71% of patients on at least one BM evaluation, and were distinct/prominent (at least visible at ×10 objective) in 57% of cases and 47% of patients. Proliferation centers were commonly seen in a perivascular or perisinusoidal distribution. Distinct/prominent PCs were more commonly found in follow-up BM biopsies than diagnostic biopsies. Although PCs were more commonly seen in our follow-up biopsies, it remains to be determined whether PCs in the BM are truly an acquired finding in the natural history of the disease. Because our cohort contained few primary diagnostic marrows, a larger study with more such cases is needed to answer this specific question. To complicate matters further, we observed that the extent of PCs may fluctuate and they were not invariably present in multiple follow-up biopsies from the same patients (data not presented). As we were unable to collect treatment histories or indications for BM evaluation on our cohort, a limitation of our study, it is uncertain whether this was related to treatment effect or simply represented a sampling artifact. It should be noted that our cohort likely represents patients with more aggressive/ treatment refractory disease, given our tertiary care center status, and therefore our data likely show a referral bias. There are a number of well-accepted prognostic factors in CLL/SLL. Patients with high Rai stage, advanced age, elevated LDH, diffuse large B-cell transformation, deletion 17p, 11q, or 6q, CD38 expression, ZAP70 expression, diffuse pattern of BM involvement, and IgH variable region unmutated status have poorer prognoses, whereas patients with
Table 3 Comparison of cytogenetic findings, degree of infiltration, and mean laboratory parameters between cases with typical and atypical PB cytology
del(11q) Trisomy 12 del(13q) del(17p) IgH translocation Degree of BM CLL infiltrate (%) LDH (U/L) γ-Globulin (g/dL)
Typical PB cytology (n = 43)
Atypical PB cytology (n = 13)
P
12/41 (29%) 14/40 (35%) 10/39 (26%) 12/39 (31%) 18/39 (46%) 70 245 0.63
1/13 (8%) 3/13 (23%) 9/19 (47%) 5/13 (39%) 5/13 (39%) 62 208 0.53
.15 .51 .01 .74 .75 .27 .23 .41
J.C. Chang et al. / Annals of Diagnostic Pathology 25 (2016) 15–19
low Rai stage, younger age, 13q abnormalities, IgH variable region mutated status, and lack of CD38 and ZAP-70 expression show better outcomes [15,18-20]. Trisomy 12 is generally regarded as an intermediate risk factor [15]. In our study, we correlated the presence/absence of PCs with many of these known prognostic factors. Consistent with the seminal CLL/SLL FISH study of Dohner et al. [18], 77% of our cohort had cytogenetic abnormalities identified by routine CLL/SLL FISH probes, and deletion 13q was a frequent finding (roughly one-third of patients). The presence of distinct/prominent PCs in the BM did not correlate with any FISH-identifiable abnormality when only the first BMs encountered on each patient were analyzed, and thus, PCs do not provide a morphologic surrogate for important prognostic cytogenetic abnormalities in our cohort. Balogh et al. [4] identified that a higher proportion of cells within lymph node PCs displayed cytogenetic abnormalities compared with the surrounding small CLL/SLL lymphocytes (by laser microdissection-based FISH), suggesting that PCs, the site of immune-mediated stimulation and active cellular proliferation, provide an environment suitable for the accumulation of genetically abnormal tumor cells. Similarly, Ciccone et al. [10] described increased frequency of 17p deletions, trisomy 12, and 14q translocations in lymph nodes rich in PCs (presumably related to the PCs) and thus a lower median survival for such histology. However, these observations were not reproducible in our study of PCs in BM. Although we did not find associations with the presence of PCs and adverse cytogenetic abnormalities, it should be noted that our patient population did contain higher frequencies of trisomy 12 and deletion 17p and a lower frequency of deletion 13q, compared with the Dohner study, highlighting a cohort biased toward CLL with more aggressive features. It is possible that PCs are observed more frequently in such cases, without a clear association between cytogenetic prognostic indicator and PC, thus explaining the high frequency of PCs in our cohort, but this requires further study. Several morphologic findings were examined in our CLL/SLL cohort and compared with known prognostic indicators. In our study, there was no correlation between the extent of BM involvement and the presence/absence of PCs. The presence of distinct/prominent PCs correlated only with the nodular pattern; there was no correlation observed between PCs and the diffuse pattern of CLL/ SLL involvement, a recognized poor prognostic infiltration pattern. There was no association between BM PCs and increased PB prolymphocytes or atypical PB morphology. This study also examined the association between PB findings and cytogenetic abnormalities. Previous studies have used various threshold values for PB findings, varying from 10% to 20% to define CLL/SLL with increased prolymphocytes or atypical cytology, with the 10% cutoff being the most common. For the sake of consistency, we chose a cutoff of 10% for PB findings. Similar to previous studies, we found that the presence of trisomy 12 was associated with increased PB prolymphocytes [2122]. We did not find any correlation between any cytogenetic abnormality and other atypical PB morphology in our patient cohort. In conclusion, this study demonstrated that PCs were a relatively frequent finding in BMs involved by CLL/SLL acquired at an academic tertiary care center. Distinct/prominent PCs were associated with a nodular BM infiltrate and a lower platelet count. Although prominent PCs were more often seen in follow-up biopsies than diagnostic biopsies, the presence or extent of PCs was not associated with other established prognostic indicators, when the first encountered BM of
19
patients was analyzed. Thus, there is no obvious prognostic significance to their presence in BMs. References [1] Foon KA, Rai KR, Gale RP. Chronic lymphocytic leukemia: new insights into biology and therapy. Ann Intern Med 1990;113:525–39. [2] Ben-Ezra J, Burke JS, Swartz WG, Brownell MD, Brynes RK, Hill LR, et al. Small lymphocytic lymphoma: a clinicopathologic analysis of 268 cases. Blood 1989;73:579–87. [3] Ghia P, Chiorazzi N, Stamatopoulos K. Microenvironmental influences in chronic lymphocytic leukaemia: the role of antigen stimulation. J Intern Med 2008;264:549–62. [4] Balogh Z, Reiniger L, Rajnai H, Csomor J, Szepesi A, Balogh A, et al. High rate of neoplastic cells with genetic abnormalities in proliferation centers of chronic lymphocytic leukemia. Leuk Lymphoma 2011;52:1080–4. [5] Soma LA, Craig FE, Swerdlow SH. The proliferation center microenvironment and prognostic markers in chronic lymphocytic leukemia/small lymphocytic lymphoma. Hum Pathol 2006;37:152–9. [6] Swerdlow SH, Murray LJ, Habeshaw JA, Stansfeld AG. Lymphocytic lymphoma/Bchronic lymphocytic leukaemia—an immunohistopathological study of peripheral B lymphocyte neoplasia. Br J Cancer 1984;50:587–99. [7] Granziero L, Ghia P, Circosta P, Gottardi D, Strola G, Geuna M, et al. Survivin is expressed on CD40 stimulation and interfaces proliferation and apoptosis in B-cell chronic lymphocytic leukemia. Blood 2001;97:2777–83. [8] Lampert IA, Wotherspoon A, Van Noorden S, Hasserjian RP. High expression of CD23 in the proliferation centers of chronic lymphocytic leukemia in lymph nodes and spleen. Hum Pathol 1999;30:648–54. [9] Asplund SL, McKenna RW, Howard MS, Kroft SH. Immunophenotype does not correlate with lymph node histology in chronic lymphocytic leukemia/small lymphocytic lymphoma. Am J Surg Pathol 2002;26:624–9. [10] Ciccone M, Agostinelli C, Rigolin GM, Piccaluga PP, Cavazzini F, Righi S, et al. Proliferation centers in chronic lymphocytic leukemia: correlation with cytogenetic and clinicobiological features in consecutive patients analyzed on tissue microarrays. Leukemia 2012;26:499–508. [11] Gine E, Martinez A, Villamor N, Lopez-Guillermo A, Camos M, Martinez D, et al. Expanded and highly active proliferation centers identify a histological subtype of chronic lymphocytic leukemia (“accelerated” chronic lymphocytic leukemia) with aggressive clinical behavior. Haematologica 2010;95:1526–33. [12] Rozman C, Montserrat E, Rodriguez-Fernandez JM, Ayats R, Vallespi T, Parody R, et al. Bone marrow histologic pattern—the best single prognostic parameter in chronic lymphocytic leukemia: a multivariate survival analysis of 329 cases. Blood 1984;64:642–8. [13] Rozman C, Hernandez-Nieto L, Montserrat E, Brugues R. Prognostic significance of bonemarrow patterns in chronic lymphocytic leukaemia. Br J Haematol 1981;47:529–37. [14] Pangalis GA, Roussou PA, Kittas C, Mitsoulis-Mentzikoff C, Matsouka-Alexandridis P, Anagnostopoulos N, et al. Patterns of bone marrow involvement in chronic lymphocytic leukemia and small lymphocytic (well differentiated) non–Hodgkin's lymphoma. Its clinical significance in relation to their differential diagnosis and prognosis. Cancer 1984;54:702–8. [15] Muller-Hermelink HK, Montserrat E, Catovsky D, Campo E, Harris NL, Stein H. In: Swerdlow SH, Campo E, Harris NL, et al, editors. WHO classification of tumours of haematopoietic and lymphoid tissues. Lyon, France: IARC; 2008. [16] Hallek M, Cheson BD, Catovsky D, Caligaris-Cappio F, Dighiero G, Dohner H. Guidelines for the diagnosis and treatment of chronic lymphocytic leukemia: a report from the International Workshop On Chronic Lymphocytic Leukemia Updating the National Cancer Institute-Working Group 1996 guidelines. Blood 2008;111:5446–56. [17] Horna P, Olteanu H, Kroft SH, Harrington AM. Flow cytometric analysis of surface light chain expression patterns in B-cell lymphomas using monoclonal and polyclonal antibodies. Am J Clin Pathol 2011;136:954–9. [18] Dohner H, Stilgenbauer S, Benner A, et al. Genomic aberrations and survival in chronic lymphocytic leukemia. N Engl J Med 2000;343:1910–6. [19] Oscier DG, Gardiner AC, Mould SJ, Glide S, Davis ZA, Ibbotson RE, et al. Multivariate analysis of prognostic factors in CLL: clinical stage, IGVH gene mutational status, and loss or mutation of the p53 gene are independent prognostic factors. Blood 2002;100:1177–84. [20] Krober A, Seiler T, Benner A, Bullinger L, Bruckle E, Lichter P, et al. V(H) mutation status, CD38 expression level, genomic aberrations, and survival in chronic lymphocytic leukemia. Blood 2002;100:1410–6. [21] Que TH, Marco JG, Ellis J, Matutes E, Babapulle VB, Boyle S, et al. Trisomy 12 in chronic lymphocytic leukemia detected by fluorescence in situ hybridization: analysis by stage, immunophenotype, and morphology. Blood 1993;82:571–5. [22] Tabernero MD, San Miguel JF, Garcia JL, Garcia-Isidoro M, Wiegant J, Ciudad J, et al. Clinical, biological, and immunophenotypical characteristics of B-cell chronic lymphocytic leukemia with trisomy 12 by fluorescence in situ hybridization. Cytometry 1995;22:217–22.
Contents lists available at ScienceDirect
Annals of Diagnostic Pathology
Proliferation centers in bone marrows involved by chronic lymphocytic leukemia/small lymphocytic lymphoma: a clinicopathologic analysis☆ Jason C. Chang, MD, Alexandra M. Harrington, MD, Horatiu Olteanu, MD, PhD., Peter VanTuinen, PhD, Steven H. Kroft, MD ⁎ Department of Pathology, Medical College of Wisconsin, Milwaukee, WI
a r t i c l e
i n f o
Keywords: Chronic lymphocytic leukemia Proliferation center Bone marrow pattern
a b s t r a c t Objectives: Proliferation centers (PCs) are a characteristic finding in chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) lymph nodes, and their presence and extent in this site are not currently felt to be related to clinical course. In contrast, detailed clinicopathologic analyses of bone marrow (BM) PCs have not been previously reported. Methods: The PCs in 88 CLL/SLL BMs from 45 patients (pts) were graded (0-4) and were correlated with other morphologic, immunophenotypic, cytogenetic, and laboratory features. Results: Proliferation centers were present in 69 BMs (78%) from 32 pts (71%) and were distinct/prominent (grades 2-4) in 21 pts (47%), with the latter more commonly found in follow-up BMs (1/7 diagnostic BMs vs 49/81 follow-up BMs; P = .04). When present, PCs were most commonly graded as distinct nodules easily visible on ×10. No relationships were identified between PCs and any complete blood count parameter, serum lactate dehydrogenase or IgG levels, degree or pattern of BM involvement, blood morphology, CD38 and FMC7 expression by flow cytometry, or fluorescence in situ hybridization results, when the first encountered BM was considered for each patient. Conclusions: This represents the first detailed analysis of PCs in CLL/SLL BMs. In our tertiary center, PCs were seen frequently, in approximately three-fourths of cases. There were no statistical associations identified between PCs and cytogenetic, immunophenotypic, or other laboratory and morphologic findings. © 2016 Elsevier Inc. All rights reserved.
1. Introduction Chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/ SLL) is a clonal lymphoproliferative disorder composed of small, mature B cells in the peripheral blood (PB), bone marrow (BM), and lymph nodes. It is the most common leukemia in Western countries, representing approximately 30% of all leukemias [1]. Lymph nodes involved by CLL/SLL typically show effacement of architecture, with a diffuse or vaguely nodular pattern, and with pale, irregular nodules distributed in a darker cellular background [2]. The pale areas represent proliferation centers (PCs), which are concentrations of prolymphocytes and paraimmunoblasts, a spectrum of larger cells with more dispersed chromatin, conspicuous nucleoli, and more abundant cytoplasm [2]. Proliferation centers are thought to represent the site of Tcell–dependent and stromal cell–dependent immune responses, resulting
☆ Conflicts of interest and source of funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. ⁎ Corresponding author at: Department of Pathology, Medical College of Wisconsin, 9200 W. Wisconsin Ave, Milwaukee, WI 53226. Tel.: +1 414 805 8459; fax: +1 414 805 8735. E-mail address: [email protected] (S.H. Kroft). http://dx.doi.org/10.1016/j.anndiagpath.2016.07.011 1092-9134/© 2016 Elsevier Inc. All rights reserved.
in the selection and proliferation of clonal B cells [3]. Not surprisingly, PCs often demonstrate genetic and immunophenotypic differences compared with the surrounding small lymphocytes, including increased expression of Ki-67, CD71, MUM1/IRF4, HLA-DC, HLA-DR, CD20, and CD23, and down-regulation of IgD and CD9 [4-8]. Although some studies have not found a correlation between clinical features and the extent of PCs in lymph nodes [9], other studies suggest that the presence of confluent PCs in CLL/SLL or SLL lymph nodes was associated with a worse clinical course and/or specific cytogenetic abnormalities [2,10-11]. Bone marrow involvement by CLL/SLL may show nodular, interstitial, diffuse, or mixed patterns of infiltration [12-14]. The pattern of involvement in CLL/SLL was once regarded as an important prognostic factor, but more recent literature has described other, more powerful predictive markers, such as cytogenetics and molecular abnormalities. Although the presence of PCs, in the setting of the appropriate immunophenotype, is essentially diagnostic of CLL/SLL in the BM, it is generally considered to be uncommon in BM biopsy specimens. Despite the robust literature devoted to CLL/SLL infiltration patterns in BM and their relationship to prognosis, there are few data on the frequency of PCs in the BM; therefore, we sought to systematically study this morphologic finding in a large cohort of CLL/SLL BM biopsies. To the best of our knowledge, no previous studies have examined the presence
16
J.C. Chang et al. / Annals of Diagnostic Pathology 25 (2016) 15–19
and extent of BM PCs in correlation with cytogenetic abnormalities, immunophenotypic features, and other laboratory findings. 2. Materials and methods 2.1. Patients All patients with CLL/SLL diagnosed according to the World Health Organization 2008 criteria [15,16], who had a BM biopsy performed at Froedtert Hospital/Medical College of Wisconsin from 2006 to 2010, were reviewed. Cases with less than 5% BM involvement by CLL/SLL were excluded. Patient demographics, the complete blood count, serum lactate dehydrogenase (LDH), and IgG immunoglobulin were recorded. This study was approved by the institutional review board at the Medical College of Wisconsin. 2.2. Morphologic studies The hematoxylin and eosin–stained histologic sections of the BM biopsy and clot section were reviewed. The BM PCs were graded together by 2 of the authors (JCC and SHK) as follows: 0, absent; 1, present, but small and ill-defined, not visible on ×10 objective; 2, distinct, easily recognized on ×10 objective; 3, extensive; and 4, diffuse increase in prolymphocytes without discrete PC formation. Absent/ill-defined PCs equated to grades 0 to 1, and distinct/prominent PCs equated to grades 2 to 4. In addition, the patterns of BM infiltrate were classified into nodular, diffuse, interstitial, and mixed by previously described definitions [12-14]. In cases where more than one pattern existed, the primary and secondary patterns were noted. The degree of BM involvement by the CLL/SLL infiltrate was semiquantitatively classified into 4 categories: less than 25%, 26% to 50%, 51% to 75%, and more than 75%. If present, the location of the PCs was noted (ie, perisinusoidal, perivascular). The PB lymphocyte morphology was assessed by reviewing the corresponding Wright-stained PB smears. Cases were considered to have increased prolymphocytes if they comprised more than 10% of the lymphocytes. Cases were categorized as atypical morphology if more than 10% of lymphocytes show cleaved or irregular nuclear contours. Correlations with concomitant lymph node biopsies could not be performed as too few cases had such concomitant biopsies. 2.3. Flow cytometric analysis Flow cytometric studies were performed using previously described methods [17]. Briefly, specimens were analyzed by 4-color flow cytometry on a FACSCanto flow cytometer with FACSDiva software (Becton Dickinson, San Jose, CA). Antibodies relevant to the current analysis included CD38 and FMC7 (Becton Dickinson, Franklin Lakes, NJ). Isotype controls were performed in all cases. Cluster analysis was performed with Paint-A-Gate software (Verity House, Topsham, ME). CD38 and FMC7 expression was assessed by the degree of overlap between the neoplastic cluster and the same population in the isotype control tube, using a 20% cutoff at a 2% isotypic control threshold. 2.4. Fluorescence in situ hybridization A CLL/SLL fluorescence in situ hybridization (FISH) panel, consisting of a commercially available set of 5 probes (Vysis, Des Plaines, IL) to evaluate for trisomy 12, IgH rearrangements, and deletions 11q22, 13q14, and 17p13, was performed according to manufacturer's instructions. Two hundred interphase nuclei were analyzed for each probe. 2.5. Statistical analysis The histopathologic characteristics of the BM PCs were correlated with other morphologic, immunophenotypic, FISH, and laboratory
findings. The associations between 2 categorical variables were examined using the Fisher exact test. Comparisons between the means of continuous variables were performed using the t test. When examining for trends across unique patients, findings from the first BM available for review were considered in the analysis. Statistical significance was set at P b .05. Statistical analyses were performed using SPSS for Macintosh version 11 (SPSS, Chicago, IL) software.
3. Results 3.1. Cases The study population consisted of 88 BM examinations from 45 patients. The patients included 37 men and 8 women ranging from 48 to 81 years (median age, 63 years). Seven cases were diagnostic BMs, whereas the remaining 81 were follow-up assessments.
3.2. Morphologic features Histologically, PCs were present in 69 (78%) of 88 cases and 32 (71%) of 45 patients in the first BMs from each available for review. The PCs in BMs were histologically identical to their counterparts in lymph nodes, appearing as pale staining, vaguely nodular collections of prolymphocytes, paraimmunoblasts, and small lymphocytes that were noticeable against a background of dark staining typical small, round CLL/SLL lymphocytes. Nineteen (22%) cases had grade 1 PCs (Fig. 1A and B). Distinct/prominent PCs (grades 2-4) were relatively common and were observed in 50 (57%) of 88 cases and 21 (47%) of 45 patients at first BM review. Grade 2 PCs were the most common pattern, observed in 38 (43%) of 88 cases, whereas grade 3 and grade 4 PCs were present in 7 (8%) and 5 (6%) of 88 cases, respectively (Fig. 1C-F). The PCs were observed adjacent to marrow sinusoids in 13 (15%) of 88 cases and in a perivascular distribution in 14 (16%) of 88 cases (Fig. 2). (See Fig. 2.) Distinct/prominent PCs were seen in 1 (14%) of 7 diagnostic BMs and 49 (60%) of 81 follow-up BMs (P = .04). However, PCs were not a temporally consistent feature in patients. Many patients with multiple follow-up biopsies had some biopsies showing distinct/prominent PCs and others with complete absence of PCs (data not shown). The most common pattern of BM involvement was the mixed pattern, seen in 58 (66%) of 88 cases, followed by interstitial (23/88; 26%), diffuse (5/88; 6%), and nodular patterns (2/88; 2%). Among cases with a mixed pattern, the most common primary pattern was the interstitial pattern (40/58 cases; 69%) and the most common secondary pattern was the nodular pattern (35/58 cases; 69%). Overall, interstitial, nodular, and diffuse patterns were observed in 81 (92%) of 88, 49 (56%) of 88, and 16 (18%) of 88 cases, respectively, either as the sole pattern or as the primary or secondary pattern in a mixed pattern. Nodular and diffuse patterns did not coexist in any given case; whenever there was a mixed pattern with either the nodular or diffuse pattern, the other component was always the interstitial pattern. Distinct/prominent PCs were seen in 33 (66%) of 50 cases containing a nodular pattern compared with those cases with absent/ill-defined PCs 16 (42%) of 38 (P = .03; Table 1), but this association disappeared when first available BMs were analyzed for each patient (P = .377). In contrast, a diffuse or interstitial pattern of infiltrate was not associated with the presence of distinct/prominent PCs in total cases (both P = 1.00; Table 1) or first available BMs (P = .729, .421). When the degree of BM involvement by CLL/SLL was semiquantitatively assessed, 11 (13%), 13 (15%), 28 (32%), and 36 (41%) of 88 of cases showed less than 25%, 26% to 50%, 51% to 75%, and more than 75% involvement, respectively. However, the degree of involvement did not correlate with the presence of distinct/prominent PCs (P = .62).
J.C. Chang et al. / Annals of Diagnostic Pathology 25 (2016) 15–19
17
Fig. 1. Morphologic classification of PCs in BM (hematoxylin and eosin). (A) Small/ill-defined PC (grade 1; ×50). (B) Small/ill-defined PC (grade 1; ×100). (C) Extensive PCs (grade 3; ×20). (D) Grade 3 PC (×50). (E) Diffuse increase in prolymphocytes without PC formation (grade 4; ×50). (F) Diffuse prolymphocytosis (grade 4; ×100).
3.3. Fluorescence in situ hybridization Complete FISH panels were performed on 43 patients on a total of 81 BMs; 1 additional patient had FISH performed only for del(11q) and trisomy [12]. Cytogenetic abnormalities were detected by FISH in 32 (71.1%) of 44 patients and 64 (79%) of 81 cases. IgH abnormalities and deletion 13q14 were the most common cytogenetic abnormalities, each present in 16 (37%) of 43 patients. The remaining cytogenetic abnormalities, in order of frequency, included trisomy 12 in 11 (26%), del 17p13 (p53) in 10 (23%), and deletion 11q22 (ATM) in 9 (21%) of 43 patients.
Fig. 2. Distribution of PCs in BM (hematoxylin and eosin). (A) Perisinusoidal PC (×20). (B) Perivascular PC (×50).
When considering all BMs with FISH (n = 81), distinct/prominent PCs were associated with an absence of trisomy 12 (Table 1), but this relationship disappeared when only analyzing the first BM of the patients (P = .175). The presence of distinct/prominent PCs was not significantly associated with other FISH abnormalities for total BMs (Table 1) or unique patients. 3.4. Immunophenotyping CD38 and FMC7 expressions were assessed in 47 and 69 cases by flow cytometry, respectively. This immunophenotypic data are presented
Fig. 3. The PB morphology examples (Wright Giemsa). (A) Case with increased PB prolymphocytes (×100). (B) Case with atypical PB cytology (×100).
18
J.C. Chang et al. / Annals of Diagnostic Pathology 25 (2016) 15–19
Table 1 Comparison of the BM pattern of CLL infiltration, cytogenetic and immunophenotypic findings, and mean clinical and laboratory parameters between BMs with absent/ill-defined PCs and BMs with prominent/distinct PCs
BM infiltration pattern Nodular pattern Diffuse pattern Interstitial pattern FISH findings del(11q) Trisomy 12 del(13q) del(17p) IgH Immunophenotype CD38+ FMC-7+ Clinical/laboratory parameters Age (y) Hemoglobin (g/dL) Platelet count (thousand/μL) Lymphocyte count (thousand/μL) γ-Globulin (g/dL) LDH (U/L)
Absent/ill-defined PCs (n = 38)
Distinct/prominent PCs (n = 50)
P
16/38 (42%) 7/38 (18%) 35/38 (92%)
33/50 (66%) 9/50 (18%) 46/50 (92%)
.03 1.00 1.00
8/36 (22%) 14/36 (39%) 10/36 (28%) 8/36 (22%) 16/36 (36%)
7/45 (16%) 7/45 (16%) 18/45 (40%) 14/45 (31%) 19/45 (42%)
.57 .02 .34 .46 1.00
12/20 (60%) 3/35 (9%)
22/27 (81%) 0/34 (0%)
.27 .24
62.2 11.3 127
62.0 11.1 90
.87 .75 .02
29.1
13.3
.08
0.65 238
0.55 224
.21 .46
in Table 1. No statistically significant relationships were identified between CD38 or FMC7 expression and PCs.
3.5. Clinical and laboratory data Of the clinical and laboratory data, only platelet count was significantly associated with the presence of distinct/prominent PCs in BMs. Patients whose BMs showed distinct/prominent PCs had a mean platelet count of 90 × 103/μL, compared with 127 × 10 3/μL in those patients whose BMs showed absent/ill-defined PCs (P = .02; Table 1); however, this relationship was not present when analyzing only the first available BM from each patient (P = .347). No other clinical and laboratory data were significantly correlated with the presence of distinct/prominent PCs in BMs (Table 1).
3.1. Peripheral blood findings Fifty-six cases had corresponding PB smears available for review. Fifteen (27%) of 56 cases showed more than 10% prolymphocytes in the PB (Fig. 3A). Increased PB prolymphocytes were present in 6 (20%) of 30 cases of BMs with distinct/prominent PCs and 9 (35%) of 26 BMs with absent/indistinct PCs (P = .24). Increased PB prolymphocytes were strongly associated with the presence of trisomy 12 (P b .01) but no other cytogenetic abnormalities (Table 2). Increased PB prolymphocytes were associated with more extensive degree of CLL BM infiltration and higher LDH values (Table 2). Thirteen (23%) of 56 cases showed atypical PB morphology (Fig. 3B). Atypical PB morphology was present in 10 (33%) of 30 cases of BMs with distinct/prominent PCs and 3 (12%) of 26 cases of BMs with absent/indistinct PCs (P = .06), but this trend was not evident when the analysis was restricted to first encountered BMs (P = .157). Atypical PB cytology was associated with deletion 13q14 (P = .01), although this association was also not evident on analysis of first encountered BMs (P = .173; Table 3). No statistically significant associations were found between atypical PB morphology and other cytogenetic abnormalities or clinical parameters (Table 3).
Table 2 Comparison of cytogenetic findings, degree of infiltration, and mean laboratory parameters between cases with and without increased PB prolymphocytes
del(11q) Trisomy 12 del(13q) del(17p) IgH translocation Degree of BM CLL infiltrate (%) LDH (U/L) γ-Globulin (g/dL)
Without increased PB prolymphocytes (n = 41)
Increased PB prolymphocytes (n = 15)
P
11/39 (28%) 5/38 (13%) 17/38 (45%) 12/38 (32%) 17/38 (45%) 63
2/15 (13%) 12/15 (80%) 2/14 (14%) 5/14 (36%) 6/14 (43%) 84
.31 b.01 .06 1.00 1.00 .001
210 0.60
312 0.62
b.001 .85
4. Discussion Although the clinical significance of PCs in CLL/SLL lymph node biopsies has been previously studied, no systematic studies to date have examined the frequency of PCs in BMs and the significance of this marrow finding. For that reason, this study aimed to assess the presence of PCs in the BM and correlate the extent of PCs to other morphologic, immunophenotypic, cytogenetic, and laboratory findings that are well established as prognostic indicators in CLL/SLL. The present study demonstrates that PCs are a relatively frequent feature in BMs involved by CLL/SLL. In our study, PCs were present in 78% of marrows overall and 71% of patients on at least one BM evaluation, and were distinct/prominent (at least visible at ×10 objective) in 57% of cases and 47% of patients. Proliferation centers were commonly seen in a perivascular or perisinusoidal distribution. Distinct/prominent PCs were more commonly found in follow-up BM biopsies than diagnostic biopsies. Although PCs were more commonly seen in our follow-up biopsies, it remains to be determined whether PCs in the BM are truly an acquired finding in the natural history of the disease. Because our cohort contained few primary diagnostic marrows, a larger study with more such cases is needed to answer this specific question. To complicate matters further, we observed that the extent of PCs may fluctuate and they were not invariably present in multiple follow-up biopsies from the same patients (data not presented). As we were unable to collect treatment histories or indications for BM evaluation on our cohort, a limitation of our study, it is uncertain whether this was related to treatment effect or simply represented a sampling artifact. It should be noted that our cohort likely represents patients with more aggressive/ treatment refractory disease, given our tertiary care center status, and therefore our data likely show a referral bias. There are a number of well-accepted prognostic factors in CLL/SLL. Patients with high Rai stage, advanced age, elevated LDH, diffuse large B-cell transformation, deletion 17p, 11q, or 6q, CD38 expression, ZAP70 expression, diffuse pattern of BM involvement, and IgH variable region unmutated status have poorer prognoses, whereas patients with
Table 3 Comparison of cytogenetic findings, degree of infiltration, and mean laboratory parameters between cases with typical and atypical PB cytology
del(11q) Trisomy 12 del(13q) del(17p) IgH translocation Degree of BM CLL infiltrate (%) LDH (U/L) γ-Globulin (g/dL)
Typical PB cytology (n = 43)
Atypical PB cytology (n = 13)
P
12/41 (29%) 14/40 (35%) 10/39 (26%) 12/39 (31%) 18/39 (46%) 70 245 0.63
1/13 (8%) 3/13 (23%) 9/19 (47%) 5/13 (39%) 5/13 (39%) 62 208 0.53
.15 .51 .01 .74 .75 .27 .23 .41
J.C. Chang et al. / Annals of Diagnostic Pathology 25 (2016) 15–19
low Rai stage, younger age, 13q abnormalities, IgH variable region mutated status, and lack of CD38 and ZAP-70 expression show better outcomes [15,18-20]. Trisomy 12 is generally regarded as an intermediate risk factor [15]. In our study, we correlated the presence/absence of PCs with many of these known prognostic factors. Consistent with the seminal CLL/SLL FISH study of Dohner et al. [18], 77% of our cohort had cytogenetic abnormalities identified by routine CLL/SLL FISH probes, and deletion 13q was a frequent finding (roughly one-third of patients). The presence of distinct/prominent PCs in the BM did not correlate with any FISH-identifiable abnormality when only the first BMs encountered on each patient were analyzed, and thus, PCs do not provide a morphologic surrogate for important prognostic cytogenetic abnormalities in our cohort. Balogh et al. [4] identified that a higher proportion of cells within lymph node PCs displayed cytogenetic abnormalities compared with the surrounding small CLL/SLL lymphocytes (by laser microdissection-based FISH), suggesting that PCs, the site of immune-mediated stimulation and active cellular proliferation, provide an environment suitable for the accumulation of genetically abnormal tumor cells. Similarly, Ciccone et al. [10] described increased frequency of 17p deletions, trisomy 12, and 14q translocations in lymph nodes rich in PCs (presumably related to the PCs) and thus a lower median survival for such histology. However, these observations were not reproducible in our study of PCs in BM. Although we did not find associations with the presence of PCs and adverse cytogenetic abnormalities, it should be noted that our patient population did contain higher frequencies of trisomy 12 and deletion 17p and a lower frequency of deletion 13q, compared with the Dohner study, highlighting a cohort biased toward CLL with more aggressive features. It is possible that PCs are observed more frequently in such cases, without a clear association between cytogenetic prognostic indicator and PC, thus explaining the high frequency of PCs in our cohort, but this requires further study. Several morphologic findings were examined in our CLL/SLL cohort and compared with known prognostic indicators. In our study, there was no correlation between the extent of BM involvement and the presence/absence of PCs. The presence of distinct/prominent PCs correlated only with the nodular pattern; there was no correlation observed between PCs and the diffuse pattern of CLL/ SLL involvement, a recognized poor prognostic infiltration pattern. There was no association between BM PCs and increased PB prolymphocytes or atypical PB morphology. This study also examined the association between PB findings and cytogenetic abnormalities. Previous studies have used various threshold values for PB findings, varying from 10% to 20% to define CLL/SLL with increased prolymphocytes or atypical cytology, with the 10% cutoff being the most common. For the sake of consistency, we chose a cutoff of 10% for PB findings. Similar to previous studies, we found that the presence of trisomy 12 was associated with increased PB prolymphocytes [2122]. We did not find any correlation between any cytogenetic abnormality and other atypical PB morphology in our patient cohort. In conclusion, this study demonstrated that PCs were a relatively frequent finding in BMs involved by CLL/SLL acquired at an academic tertiary care center. Distinct/prominent PCs were associated with a nodular BM infiltrate and a lower platelet count. Although prominent PCs were more often seen in follow-up biopsies than diagnostic biopsies, the presence or extent of PCs was not associated with other established prognostic indicators, when the first encountered BM of
19
patients was analyzed. Thus, there is no obvious prognostic significance to their presence in BMs. References [1] Foon KA, Rai KR, Gale RP. Chronic lymphocytic leukemia: new insights into biology and therapy. Ann Intern Med 1990;113:525–39. [2] Ben-Ezra J, Burke JS, Swartz WG, Brownell MD, Brynes RK, Hill LR, et al. Small lymphocytic lymphoma: a clinicopathologic analysis of 268 cases. Blood 1989;73:579–87. [3] Ghia P, Chiorazzi N, Stamatopoulos K. Microenvironmental influences in chronic lymphocytic leukaemia: the role of antigen stimulation. J Intern Med 2008;264:549–62. [4] Balogh Z, Reiniger L, Rajnai H, Csomor J, Szepesi A, Balogh A, et al. High rate of neoplastic cells with genetic abnormalities in proliferation centers of chronic lymphocytic leukemia. Leuk Lymphoma 2011;52:1080–4. [5] Soma LA, Craig FE, Swerdlow SH. The proliferation center microenvironment and prognostic markers in chronic lymphocytic leukemia/small lymphocytic lymphoma. Hum Pathol 2006;37:152–9. [6] Swerdlow SH, Murray LJ, Habeshaw JA, Stansfeld AG. Lymphocytic lymphoma/Bchronic lymphocytic leukaemia—an immunohistopathological study of peripheral B lymphocyte neoplasia. Br J Cancer 1984;50:587–99. [7] Granziero L, Ghia P, Circosta P, Gottardi D, Strola G, Geuna M, et al. Survivin is expressed on CD40 stimulation and interfaces proliferation and apoptosis in B-cell chronic lymphocytic leukemia. Blood 2001;97:2777–83. [8] Lampert IA, Wotherspoon A, Van Noorden S, Hasserjian RP. High expression of CD23 in the proliferation centers of chronic lymphocytic leukemia in lymph nodes and spleen. Hum Pathol 1999;30:648–54. [9] Asplund SL, McKenna RW, Howard MS, Kroft SH. Immunophenotype does not correlate with lymph node histology in chronic lymphocytic leukemia/small lymphocytic lymphoma. Am J Surg Pathol 2002;26:624–9. [10] Ciccone M, Agostinelli C, Rigolin GM, Piccaluga PP, Cavazzini F, Righi S, et al. Proliferation centers in chronic lymphocytic leukemia: correlation with cytogenetic and clinicobiological features in consecutive patients analyzed on tissue microarrays. Leukemia 2012;26:499–508. [11] Gine E, Martinez A, Villamor N, Lopez-Guillermo A, Camos M, Martinez D, et al. Expanded and highly active proliferation centers identify a histological subtype of chronic lymphocytic leukemia (“accelerated” chronic lymphocytic leukemia) with aggressive clinical behavior. Haematologica 2010;95:1526–33. [12] Rozman C, Montserrat E, Rodriguez-Fernandez JM, Ayats R, Vallespi T, Parody R, et al. Bone marrow histologic pattern—the best single prognostic parameter in chronic lymphocytic leukemia: a multivariate survival analysis of 329 cases. Blood 1984;64:642–8. [13] Rozman C, Hernandez-Nieto L, Montserrat E, Brugues R. Prognostic significance of bonemarrow patterns in chronic lymphocytic leukaemia. Br J Haematol 1981;47:529–37. [14] Pangalis GA, Roussou PA, Kittas C, Mitsoulis-Mentzikoff C, Matsouka-Alexandridis P, Anagnostopoulos N, et al. Patterns of bone marrow involvement in chronic lymphocytic leukemia and small lymphocytic (well differentiated) non–Hodgkin's lymphoma. Its clinical significance in relation to their differential diagnosis and prognosis. Cancer 1984;54:702–8. [15] Muller-Hermelink HK, Montserrat E, Catovsky D, Campo E, Harris NL, Stein H. In: Swerdlow SH, Campo E, Harris NL, et al, editors. WHO classification of tumours of haematopoietic and lymphoid tissues. Lyon, France: IARC; 2008. [16] Hallek M, Cheson BD, Catovsky D, Caligaris-Cappio F, Dighiero G, Dohner H. Guidelines for the diagnosis and treatment of chronic lymphocytic leukemia: a report from the International Workshop On Chronic Lymphocytic Leukemia Updating the National Cancer Institute-Working Group 1996 guidelines. Blood 2008;111:5446–56. [17] Horna P, Olteanu H, Kroft SH, Harrington AM. Flow cytometric analysis of surface light chain expression patterns in B-cell lymphomas using monoclonal and polyclonal antibodies. Am J Clin Pathol 2011;136:954–9. [18] Dohner H, Stilgenbauer S, Benner A, et al. Genomic aberrations and survival in chronic lymphocytic leukemia. N Engl J Med 2000;343:1910–6. [19] Oscier DG, Gardiner AC, Mould SJ, Glide S, Davis ZA, Ibbotson RE, et al. Multivariate analysis of prognostic factors in CLL: clinical stage, IGVH gene mutational status, and loss or mutation of the p53 gene are independent prognostic factors. Blood 2002;100:1177–84. [20] Krober A, Seiler T, Benner A, Bullinger L, Bruckle E, Lichter P, et al. V(H) mutation status, CD38 expression level, genomic aberrations, and survival in chronic lymphocytic leukemia. Blood 2002;100:1410–6. [21] Que TH, Marco JG, Ellis J, Matutes E, Babapulle VB, Boyle S, et al. Trisomy 12 in chronic lymphocytic leukemia detected by fluorescence in situ hybridization: analysis by stage, immunophenotype, and morphology. Blood 1993;82:571–5. [22] Tabernero MD, San Miguel JF, Garcia JL, Garcia-Isidoro M, Wiegant J, Ciudad J, et al. Clinical, biological, and immunophenotypical characteristics of B-cell chronic lymphocytic leukemia with trisomy 12 by fluorescence in situ hybridization. Cytometry 1995;22:217–22.