Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Am J Surg Pathol. Author manuscript; available in PMC 2014 March 1.
Published in final edited form as:
PMCID: PMC3566339

Follicular Lymphomas in children and young adults: A comparison of the pediatric variant with usual follicular lymphoma


Follicular lymphoma (FL), a common lymphoma in adults, occurs rarely in pediatric and young adult patients. Most pediatric cases have been described as Grade 3, but the criteria to distinguish the pediatric variant of FL (PFL) from usual FL (UFL) in adults are not well defined. We undertook a study of FL in patients under age 30. We identified 63 cases, which were analyzed by morphology, immunohistochemistry, and PCR analysis of IGH@ and IGK@ clonality. These data were correlated with clinical findings including stage, treatment, and outcome. Among the 63 cases, 34 cases were classified as PFL; 22 presenting in lymph nodes, 8 in Waldeyer’s ring and 4 in testis. Clonal immunoglobulin gene rearrangement was detected in 97% of PFL cases, but FISH analysis showed an absence of the BCL2/IGH@ translocation in all cases tested. Twenty-nine cases were classified as UFL, 28 of which presented in lymph nodes. The nodal PFL cases were observed exclusively in males in both children and young adults with a median age of 15 years. They showed marked head/neck predilection, blastoid cytological features with a high proliferation rate, lack of BCL2 protein and t(14;18), low clinical stage at presentation, and good prognosis. PFL involving Waldeyer’s ring were distinguished by MUM1 expression, 50% (3/6) of which carried IRF4 breaks. BCL2 expression was common (63%) in the absence of BCL2/IGH@ translocation. UFL cases were more common in females, exclusively in young adult patients (median age, 24 years) with no patients under age 18. 25/29 were grade 1–2, and four cases were classified as grade 3A. They exhibited a higher clinical stage at presentation. 83% expressed BCL2. Our results indicate that histological and immunophenotypic criteria can reliably separate PFL and UFL, and that UFL is exceptionally rare in the pediatric age group. PFL associated with particular anatomic sites have distinctive features, and should be evaluated separately in future clinical and biological studies.

Keywords: follicular lymphoma, follicular lymphoma, IRF4 (interferon regulatory factor 4), MUM1, BCL2, BCL6, Fluorescence in-situ hybridization (FISH), Waldeyer’s ring, testis, pediatric


Follicular lymphoma (FL) is the second most common lymphoma worldwide. It mainly affects adults and rarely occurs in the pediatric population and young adult patients.1 Pediatric follicular lymphoma (PFL) is a newly recognized variant of FL included in the 2008 WHO Classification of Tumors of Hematopoietic and Lymphoid Tissues.1 Limited studies have reported on the histologic, molecular, and clinical features of FL in young patients. 28. PFL is reported to be characterized by high histologic grade, and the lack of BCL2 protein expression and t(14;18). Clinically, PFL predominantly occurs in males, often presents with localized disease, and carries good prognosis. (11–13) Currently, the pathogenesis of this entity is largely unknown, and differences between nodal PFL and PFL in other anatomic sites have not been explored. While differences in cytological grade have been noted with usual FL (UFL) as seen in adults, PFL is often a diagnostic challenge, and more precise histologic criteria have not been defined.

Although a conservative approach towards FL in pediatric patients had been proposed by Atra et 1998, 2 the optimal clinical management remains unclear. Furthermore, clinicopathologic features of FL in young adult patients have not been extensively studied. In this study, we examined the histologic features, immunophenotypic profiles, immunoglobulin gene rearrangement by PCR, cytogenetic characteristics by FISH, as well as the clinical follow up data of 63 FL cases in pediatric (<= 18 year-old), and young adult (19–29 year-old) patients. We used both the WHO criteria and new observations to separate PFL from UFL in the young adult population, and to better delineate the clinical, histologic and immunophenotypic spectrum of PFL. These results will be of value in guiding the diagnosis and management of FL and the PFL variant.


Case selection

An electronic search of the pathology data base at the National Cancer Institute (NCI), National Institutes of Health (NIH), Bethesda, MD, was conducted for the diagnosis of FL, or consistent with FL, restricted to age < 30, from 2000 to 2011. Sixty eight cases with material available for review were identified. Five cases were excluded for the following reasons: 1) revised diagnosis of follicular hyperplasia (1 case); 2) revised diagnosis of marginal zone lymphoma (1 case); 3) excluded based on a component of diffuse large B-cell lymphoma (3 cases). Sixty-three cases comprised the final study cohort. Cases of nodal FL (51) were classified according to the 2008 WHO Classification of Tumors of Hematopoietic and Lymphoid Tissues as PFL or UFL, without knowledge of the patient’s age or other clinical data.1 Criteria for nodal PFL included large expansile follicles, dim to absent BCL2 expression, and a high proliferative rate. This study was approved by the Institute Review Board of the National Cancer Institute. Clinical information and follow up were obtained from submitted patients records or referring physicians in accordance with medical practice guidelines. Ten pediatric nodal marginal zone lymphoma (PNMZL) cases were reviewed concurrently in order to compare histologic features of PFL with those of PNMZL.9

Histology and Immunohistochemistry

The morphologic and immunophenotypic features were studied on formalin-fixed and paraffin-embedded (FFPE) tissue sections of the diagnostic biopsies. Immunohistochemical stains were performed using an automated immunostainer (Ventana Medical Systems, Inc, Tucson, AZ) as described previously.10 In brief, antigen retrieval was performed using a Tender Cooker (Nordicware, Minneapolis, MN) with citrate buffer. The immunohistochemical panel included CD20, CD3, CD10, BCL2 (clone 124), BCL6, MUM1, IgD, CD21, CD23, MIB1(Ki-67), and PD-1(CD279). Appropriate controls were included in all cases. BCL2 and CD10 were scored as positive if more than 50% of tumor cells exhibited staining; for BCL6 and MUM1 the required value was 30%. The E17 clone reactive with the mutated form of BCL2 (Epitomics) was examined in a subset of the cases, negative for BCL2 with clone 124 (Dako).11

PCR Analysis of Immunoglobulin Gene Rearrangement

PCR analysis of immunoglobulin gene rearrangement was performed as described previously.10 Genomic DNA was extracted from FFPE tissue sections using QIAamp DNA FFPE tissue kit (QIAGEN Inc, Valencia, CA) according to the manufacturer’s instructions. For immunoglobulin heavy chain (IGH@) clonality assays, framework 3 and framework 2 were amplified by PCR using consensus primer sets reported as Segal et al 12 and Ramasamy et al 13, respectively. For κ light chain (IGK@) clonality assays, IGK@ was amplified by PCR using ABI Fluorescence Detection Kit (In VivoScribe Technologies, INC, San Diego, CA) according to the manufacturer’s instructions. After amplification, the PCR products were analyzed using ABI 3130xl Genetic Analyzer (Applied Biosystem), and the results were interpreted based on the criteria developed in the laboratory.10 The tissue biopsies (lymph node, tonsil, testis) were extensively involved by the abnormal follicles, comprising the vast majority of the B-cells present; thus, enrichment for tumor cells was not required. In a single tonsillar case (Case 8), focal involvement was present; in this case macrodissection of the involved portion was performed to enrich for tumor DNA.

Construction of Tissue Microarray (TMA)

For PFL cases with available tissue blocks, a TMA was constructed to facilitate analysis of interphase cytogenetics, as described previously. 14 Representative tumor areas were selected based on the corresponding H&E slides and marked directly on the donor block. Two tissue cores (diameter 1.5mm) were transferred from the selected region of each donor block and arrayed into the recipient block using TMArrayer (Pathology Devices, Westminster, MD). Following the TMA construction, 5µm sections were cut with a microtome. An H&E stain was obtained to confirm inclusion of diagnostic tissue.

FISH Analysis

Interphase fluorescence in-situ hybridization (FISH) analysis for the detection of breaks or gene fusion affecting the IGH, BCL2, BCL6(LSI dual-color break apart probes and IGH-BCL2 dual color dual-fusion translocation probe, Vysis- Abbott, IL, USA), and IRF4 loci was carried out on formalin fixed paraffin embedded (FFPE) sections from tissue microarray (TMA) or whole tissue sections of the PFL cases as described previously.15 Breaks of the CBFA2T3 gene, recently described to be translocated in pediatric germinal center derived lymphomas, were also investigated using previously described probes.16 FISH analysis for the detection of IGH@, BCL6, and IRF4 breaks was performed on each case that had material available, whereas FISH analysis for BCL2 and CBFA2T3 breaks was carried out only on the cases that contained IGH@ breaks. For the cases harboring breaks in both of IGH@ and IRF4 loci, IGH-IRF4 gene fusion was confirmed by using fusion probes.16


Histologic and Immunophenotypic Findings

Thirty four of 63 FL cases were classified as PFL (Table 1), and 29 cases were classified as UFL (Table 2).

Summary of Clinical Features of PFL Patients
Summary of Clinical Features of UFL Patients


Cases diagnosed as PFL presented in lymph node (22 cases), Waldeyer’s ring (tonsils or tonsils and adenoids) (8 cases), and testis (4 cases). Differences in the histologic and immunophenotypic features were noted, depending on the anatomic site, and therefore these groups will be separately described. All cases of PFL would be considered Grade 3, based on the proportion of centroblasts/blastoid cells.

PFL involving tonsil (Fig.1) were characterized by large, closely-packed, expansile follicles composed predominantly of centroblasts (3/8 cases, Fig. 2A) or intermediate sized blastoid cells (5/8 cases, Fig. 2B). The blastoid cells had round to oval nuclei, finely clumped chromatin, small nucleoli, and scant cytoplasm. A “starry sky” appearance, a feature commonly seen in reactive follicles, was sometimes maintained. The subset involving tonsil was more likely to be positive for BCL2 protein (5/8, 63%), and notably all cases tested were positive for MUM1 (Table 3). CD10 was expressed in 63% (5/8) and BCL6 was positive in 100% (8/8) of the cases, confirming a germinal center derivation. BCL6 and CD10 positive cells were confined to the follicles. One case showed evidence of kappa light chain restriction by immunohistochemistry.

Figure 1
Histologic and immunophenotypic features of PFL in Tonsil, case 1. The architecture is focally effaced by large, expansile atypical follicles (A 20X). A few reactive follicles are seen on the left adjacent to the lesion. The neoplastic cells are positive ...
Figure 2
Comparison of cytologic Features of PFL in Tonsil and Lymph Node (1000X). Centroblasts comprised the dominant cell type in 3 of the 8 tonsillar PFL cases (A, case 1), and while the remaining 5 tonsillar PFL cases contained a predominance of monotonous, ...
Immunophenotypic Features, Clonality by PCR, and FISH Results of PFL Cases

PFL involving lymph nodes showed effaced nodal architecture by large ill-defined follicles with attenuated lymphoid cuffs. (Fig. 3) A prominent starry sky pattern was present in all cases. The follicles were often serpiginous or irregularly-shaped. Occasional nodal PFL cases showed more uniform, closely packed follicles, similar to those seen in tonsillar PFL cases. (Supplemental digital content 1) A peripheral rim of paler cells suggesting monocytoid differentiation was sometimes present, but was a minor component. The cytological composition was similar in all cases, and was dominated by monotonous, small to medium sized blastoid cells with round to oval nuclei, finely dispersed chromatin, small nucleoli, and scant cytoplasm (Fig. 2C and 2D). Typical centroblasts with peripheral membrane-bound nuclei were infrequent. Centrocyte-like cells were sometimes present, but had somewhat dispersed chromatin.

Figure 3
Histologic and immunophenotypic features of nodal PFL, case 22 (200×) A. The nodal architecture is effaced by ill-defined follicles. A starry sky pattern is evident, but without polarization characteristic of reactive germinal centers. (200 X). ...

PFL in lymph nodes were consistently and strongly positive for CD10 (100%), with CD10 highlighting the often irregular follicle centers (Fig.3). BCL6 was positive as well, with only 18% showing any expression of BCL2 protein, which was weakly positive. Eight cases of nodal PFL negative for BCL2 with clone 124 were examined with clone E17, and were negative as well. CD10 and BCL6 positive cells did not extend beyond the follicles. MUM1 was positive in 1/18 (5.5%) nodal PFL cases; this case was also positive for CD10, BCL6, CD5, and lambda light chain. Among the tonsillar or nodal MUM1 positive cases, cytologic features of plasmacytoid differentiation were not observed, even in the presence of light chain restriction by immunohistochemistry. Interfollicular B-cells were not markedly increased, helping in the distinction from PNMZL. In four cases with marginal zone differentiation, a perifollicular rim of B-cells could be identified.

PFL involving the testis contained back to back atypical small follicles composed of predominantly of centroblasts infiltrating between seminiferous tubules (Supplemental digital content 2). The follicles lacked well-defined mantle cuffs and did not show polarization. All cases were negative for BCL2 and MUM1, but were positive for BCL6 and CD10 in 4/4 (100%) and 3/4 (75%) of the cases, respectively. The follicles had a back to back arrangement, but interfollicular B-cells beyond the follicles were not present. The follicular structures in all cases of PFL were highlighted by stains for CD21 and CD23. IgD identified thin and attenuated mantle zones. Tumor cells were negative for IgD. All PFL cases showed a high (>=75%) or moderate to high (50–75%) proliferation rate assessed by MIB1 stain.


Twenty-nine cases were diagnosed as UFL, 25 classified as grade 1–2 and 4 classified as grade 3A. No cases of grade 3B were identified. All cases but one presented in lymph node; one case was diagnosed in thyroid, but the patient had disseminated disease. The histologic features were similar to those of adult UFL, in which the normal architecture was effaced or partially effaced by closely-packed atypical follicles that showed absent or attenuated mantle cuffs and lack of polarity (Supplemental digital content 3). A “starry sky” appearance was no longer present. Centrocytes comprised the majority of the cells present in the grade 1–2 UFL cases, whereas greater than 15/HPF centroblasts were present in grade 3a UFL cases. However, nodal UFL lacked the monotonous blastoid cytology that dominated nodal PFL.

All cases were positive for CD20 and BCL6. Of UFL classified as grade 1–2, 100% (21/21) were CD10 (+) and 83% (20/24) were BCL2(+). All but two grade 1–2 UFL cases showed low (<=25%) or low to moderate (25–50%) proliferation rates by MIB1 stain, with higher rates seen in two cases. Of the four grade 3A UFL cases, 2 of 4 (50%) were CD10 (+), and 2 of 3 (66.7%) cases tested were BCL2 (+). All grade 3 UFL cases showed a high proliferation rate (>75%).


Ten pediatric nodal marginal zone lymphoma cases were reviewed concurrently in order to compare the morphologic features between PFL and PNMZL (Table 4). The cases showed characteristic histologic and immunophenotypic features as described previously, 9 such as increased interfollicular B cells and fragmentation of residual germinal centers resembling progressive transformation of germinal centers (PTGC). (Supplemental digital content 4) Residual hyperplastic germinal centers maintained a starry sky appearance, showed evidence of polarization best visualized with immunohistochemistry for Ki-67, and had a heterogeneous cellular composition. The follicles were widely spaced, with less prominent CD10 positivity.

Comparison of Clinicopathologic Features of Nodal PFL, PNMZL, and UFL

PD-1 staining pattern

To further facilitate the distinction between PFL and PNMZL, staining for PD-1(CD279) was performed on PFL (24), UFL (16), and PNMZL (9) cases. Distinctive patterns of PD-1 staining in the follicles were observed among tonsillar PFL, nodal PFL, nodal UFL, and PNMZL cases. (Supplemental digital content 5). In reactive follicles PD-1 positive lymphocytes were abundant and more numerous in the light zone. This polarity was absent in the atypical follicles involved by PFL, UFL, and PNMZL. Generally, tonsillar PFL cases showed absent or a few scattered PD-1 positive cells. PD-1 positive cells were somewhat more frequent in nodal PFL, but tended to be concentrated in the periphery, giving an impression that these cells were pushed aside by the neoplastic cells. PD-1 positive cells were more abundant in the germinal centers of PNMZL, as compared with nodal PFL. UFL cases showed large numbers of PD-1 positive cells with a loss of polarity in the atypical follicles in comparison to that seen in hyperplastic follicles.

PCR and FISH Findings

A majority of the PFL cases demonstrated clonal immunoglobulin gene rearrangement by PCR analysis (Table 3). The percentage of positive cases varied slightly among different anatomic sites, but the average rate of positivity was 97%. Four cases were negative with primers to Fr2 and Fr3, but positive with primers to the Ig κ gene segments.

FISH analysis to identify breaks in the IGH@, IRF4, and BCL6 loci was performed on FFPE sections of the PFL cases with available material. FISH analysis for BCL2 and CBFA2T3 breaks was carried out only if the case had an IGH@ break. IGH@ breaks were identified in 21% (5/24) of the PFL cases, including 3 Waldeyer’s ring PFL cases, and 2 nodal PFL cases. Among the 22 PFL cases tested for IRF4 breaks, three Waldeyer’s ring cases (case 5, 7, and 8) carried IRF4 breaks. One of the cases tested also showed an IGH break, and an IGH@-IRF4 fusion was confirmed by FISH (Fig. 4). FISH analysis for IGH@ breaks could not be performed on the third case due to technical reasons. None of the PFL cases had BCL2, BCL6 or CBFA2T3 breaks.

Figure 4
FISH analysis of a MUM1-positive tonsillar PFL. IGH@-IRF4 fusion is shown using a locus-specific probe consisting of BAC clones labeled in spectrum orange (IRF4) and spectrum green (IGH@). The IGH2-IRF4 fusion is a reciprocal fusion, in which in most ...

Clinical Findings

The clinical data of the FL cases are summarized in Tables 1, ,2,2, and a comparison of the features between UFL and PFL is summarized in Table 5. The PFL cases occurred in both pediatric and young adult patients with marked male predominance (M:F = 29:5) and a median age of 14 years (age range 3–25) (Tables 1, ,5).5). All of the Waldeyer’s ring and testicular PFL cases were from pediatric patients, whereas 68% (15/22) of the nodal PFL cases were in the pediatric age group (age range 3–18), and 32% were in young adult patients (age range 19–25). Nodal PFL was seen exclusively in males, and presented predominantly in the head/neck region (77%). Four PFL cases showed a concurrent diffuse large B cell lymphoma (DLBCL) component (2 Waldeyer’s ring cases, 2 nodal cases). All PFL cases presented with localized disease and were clinical stage I (E) or II (E). The treatment regimens varied among the patients, including surgical excision only and excision with chemotherapy with or without local radiation. The components of the chemotherapy regimens also varied, but R-CHOP (R, Rituximab; CHOP, cyclophosphamide, adriamycin, vincristine, prednisone) was the most common regimen. All PFL patients achieved complete remission and showed no evidence of disease (NED) upon completion of treatment for various lengths of follow up periods. The median follow up period was 18 months (ranging from 1 to 120 months). A few patients achieved long-term disease free survival (patient #2, #9, and #25) following surgical excision only. Several recently-diagnosed PFL patients elected to be managed by surgical excision with close clinical follow up.

Comparison of Clinical Features of PFL and UFL Patients

Twenty nine of the 63 cases were classified as UFL, including 25 grade 1–2 cases and 4 grade 3A cases (Tables 2, ,5).5). The median age of this group was 24 year-old (range 18–29) with a female predominance (M:F=9:20). Interestingly, all four grade 3A UFL cases were diagnosed in females. Notably no case of UFL was diagnosed under age 18. All UFL cases but one were diagnosed in lymph node biopsies. The single case with extranodal involvement (thyroid) had disseminated disease (case 39). 74% of the patients presented with multiple sites of lymphadenopathy with or without hepatosplenomegaly and/or presented as stage III or VI disease. For the cases that had available clinical staging information, 57% presented with clinical stage III-IV disease and 38% presented with stage I disease (5/14 stage IV, 3/14 stage III, 6/14 stage I). Most patients were treated with various regimens including chemotherapy with or without radiation, or radiation only. A few patients were managed by close clinical follow up only. A majority of the patients achieved complete remission upon treatment except one patient, who showed recurrent disease 3 years after initial treatment (case 53).


In this study we analyzed the clinicopathologic features of FL in pediatric and young adult patients. Using published criteria and new observations based on our analysis of these cases, we recognized two groups, usual FL (UFL) resembling FL as seen in older adults, and PFL, lesions with distinctive histologic, immunophenotypic and molecular features. Clinically, the UFL cases were seen exclusively in young adult patients, with no case diagnosed under age 18. Patients with UFL showed a female predominance and had more advanced clinical stage at presentation. In contrast, the nodal PFL cases were seen in both pediatric and young adult patients, exclusively in males, with a predilection for the head/neck region and low clinical stage at presentation. All of the PFL patients achieved complete remission following various treatment regimens, including surgical excision alone. These data suggest that PFL is a localized disease with good prognosis. For these patients, “Watch and wait” may be the best course of action after surgical excision, as previously suggested.2

PFL and UFL show significant differences in clinical presentation, and appear to differ in clinical outcome. Therefore, correct pathologic diagnosis is essential in guiding the optimal clinical management. In addition to the recognized architectural and cytological features of nodal PFL delineated in the WHO classification (broad expansile follicles, high histological grade), our study highlighted more subtle differences in cytological composition. While PFL cases have been generally considered Grade 3 based on a high proliferation rate, we observed that the neoplastic cells in nodal PFL were neither typical centrocytes nor centroblasts in most cases. The cells were often medium in size, monotonous in cytological composition, with dispersed chromatin and inconspicuous nucleoli. This blastoid appearance correlated with a high proliferative rate. Understandably these cases have been categorized as grade 3, but diagnosis is facilitated by the appreciation of these distinctive cytological features.

At times, making the distinction between nodal PFL and florid follicular hyperplasia with monotypic light chain expression and clonal immunoglobulin gene rearrangement by PCR, 17 can be extremely difficult. The retained “starry sky” appearance characteristically seen in nodal PFL cases further complicates the matter. In these situations, recognition of a monotonous, blastoid cytological composition without polarization, or typical centroblasts and centrocytes is helpful. The extent of the architectural alteration is also useful. PCR for detection of a clonal B-cell population appears to be a very sensitive test in PFL, a finding enhanced by the use of multiple primer sets.18 Four cases negative for clonality by PCR with primers to the immunoglobulin heavy chain genes were positive when kappa probes were also employed.

A notable observation was uniform positivity for MUM1 in all evaluable cases of PFL involving Waldeyer’s ring. In contrast, MUM1 was expressed in only 5.5% of nodal PFL and no case of testicular PFL. All of the MUM1 positive cases expressed BCL6, and 57% expressed CD10 (4/7, Table 6). BCL2 protein was expressed in 63% of the tonsillar cases as well. Prominent cytologic features of plasmacytic differentiation were not evident on H&E sections, and restricted light chain expression was rarely observed.

Immunophenotypic Features and FISH Results of MUM1+ PFL Cases

In view of the frequent expression of MUM1, we were interested in examining PFL for breaks involving IRF4/MUM1. Of the Waldeyer’s ring cases tested, IRF4 breaks were detected in 3 of 6 cases, and an IGH/IRF4 fusion was confirmed in 1 of 2 cases tested. These data suggest that while IGH@/IRF4 fusion may represent one possible mechanism for the up-regulation of MUM1 expression, other mechanisms probably also exist. Recent studies have highlighted the significance of MUM1 expression in B-cell lymphomas in children and young adults. Salaverria et al.15 identified recurrent IGH@/IRF4 fusions in both diffuse large B-cell lymphomas of the germinal center B-cell type (GCB), and grade 3 FL. Sixteen of the twenty positive cases in the published study were diagnosed in patients younger than age 30, and with frequent involvement of Waldeyer’s ring (80%) as in our study. Both series identified coexpression of MUM1 with BCL6, which is an aberrant immunophenotype as these proteins are reciprocally expressed in normal germinal centers.19 Salaverria et al. also reported that the cases harboring an IGH@/IRF4 translocation had frequent alterations in the BCL6 locus, including breaks and mutations.15 While we did not identify chromosomal breaks in the BCL6 gene in any of our PFL cases, we did not perform sequence analysis to determine if there were mutations in the BCL6 gene resulting in disruption of the BCL6-IRF4-PRDM1 regulatory pathway.

Originally identified by cloning the chromosomal breakpoints of a multiple myeloma cell line with t(6;14), the IRF4 gene is a member of the interferon-regulatory factor family of transcriptional factors. MUM1 shows a biphasic expression pattern in B cells, i.e., it is expressed in immature B cells in bone marrow, absent in centroblasts of the germinal centers, and is again expressed in a subset of centrocytes in the light zone of germinal centers, and plasma cells. 14,20 It is generally believed that the expression of IRF4/MUM1 and BCL6 in B cells is mutually exclusive in normal germinal centers, but the positive cases of PFL aberrantly co-expressed both proteins. 21 UFL typically lacks MUM1 expression.1 However, the reported positive cases are usually CD10 negative, of high histologic grade, clinically more aggressive, and interestingly occur in an older patient population. 22

In comparison to PFL in other anatomic sites, several notable differences were observed in PFL involving Waldeyer’s ring. In addition to uniform positivity for MUM1, these cases were more often BCL2 positive (63%), had a higher frequency of IGH@ breaks (50%), and lacked a male predominance. These data suggest that PFL of Waldeyer’s ring may have a different molecular pathogenesis from other forms of PFL.

As expected, the nodal PFL cases in our series were generally negative for BCL2 protein (18% positivity and generally weak). These data are consistent with the previously published studies in which BCL2 positivity was reported to be 25%–55%. 3,4 Although advanced disease at presentation and resistance to combination chemotherapy have been reported to be associated with BCL2 expression, 3 a relationship between BCL2 expression and clinical stage or outcome was not observed in our study. The PFL patients in our series consistently presented with early clinical stage (I or II) and all achieved complete remission without recurrence during the available clinical follow up. Our data are in keeping with another recently published study.4 Earlier studies also have demonstrated the lack of t(14;18) in PFL, 3,4 a finding confirmed in our series.

The distinction between PFL and PNMZL also can be problematic, given the overlapping features of these two entities. However, characteristic features seen in PNMZL, 9 such as expanded interfollicular areas by B cells and PTGC-like changes, were usually absent in PFL cases. Residual hyperplastic germinal centers may be present, and show the expected cytological composition and polarization. PD-1(programmed death-1), a member of the CD28 costimulatory receptor family, is expressed by follicular helper T cells (TFH). 23 We sought to use the PD-1 staining pattern to help distinguish PFL and PNMZL. Overall, the abnormal follicles of nodal PFL contained many fewer PD-1 positive T-cells than the disrupted follicles of the PNMZL cases. Additionally, the distribution of the PD-1 positive lymphocytes was abnormal in nodal PFL, being pushed to the peripheral rim of the atypical follicles. Muenst et al. also studied the distribution of PD-1 positive lymphocytes in B-cell lymphomas.24 They observed fewer PD-1 positive lymphocytes in Grade 3 FL as compared with Grade 1–2 FL, in keeping with our observations, but their study did not include nodal MZL.

Utilizing the published and expanded criteria for the diagnosis of PFL and UFL, lymph nodes involved by FL in patients under the age of 30 clearly segregate into two groups. UFL was not observed under age 18, and is much more likely to present with disseminated disease. In addition, the data in our study indicate substantial differences in PFL involving Waldeyer’s ring from those presenting in lymph nodes or testis. Up regulation of MUM1 expression and/or IRF4 breaks imply a possible pathogenetic mechanism for PFL located in Waldeyer’s ring. A prior publication 25 had emphasized the unique features of testicular FL, which rarely if ever show evidence of extratesticular disease.

Correct diagnosis of FL in these patients requires careful examination of histologic, immunophenotypic, cytogenetic features, and clinical features, but the optimal clinical management is still unresolved. While monoclonality is demonstrated for these lesions, many cases appear to have a very limited potential for progressive disease. There is an increasing recognition in the literature of clonal lymphoid proliferations with limited potential for clinical aggressiveness.26 In the future nodal and testicular PFLs may be grouped with other clonal B-cell proliferations of undetermined significance, such as monoclonal gammopathy of undetermined significance or MGUS, follicular lymphoma in situ, and mantle cell lymphoma in situ.26,27 However, as PFL involving Waldeyer’s ring are associated with a recurrent cytogenetic abnormality, they may have potential for a more aggressive clinical course.

Supplementary Material


Supplemental Digital Content 1:

A nodal PFL case with large expansile atypical follicles and diffuse MUM1 expression, case 19 (20X). The nodal architecture is effaced by large, expansile atypical follicles (A). High power view (B, 400X) showed that the neoplastic cells are predominantly medium sized blastoid cells with some tingible body macrophages. The neoplastic cells are positive for CD20 (C), CD10 (D), BCL6 (E), and MUM1 (F). MIB1 indicates a high proliferation rate (G); CD3 (H) highlighted the T cells in the periphery of the follicles.


Supplemental Digital Content 2:

Testicular PFL, case 12. The testicular architecture is largely replaced by closely packed, rather uniform, small atypical follicles (A, 40X) which are predominantly composed of centroblasts (B, 400X). The cellular composition is more heterogeneous than that seen in the tonsillar or nodal PFL cases, with admixed centrocytes clearly evident.


Supplemental Digital Content 3:

Histologic and immunophenotypic features of nodal UFL, grade 1–2 (case 35, A, B, and C) and grade 3A (case 60, D, E, and F). These cases show characteristic features of UFL similar to those seen in adults. The nodal architecture is effaced by closely packed, rather uniform sized follicles which lack well-defined mantle cuffs, loss of polarity, and acomposed of predominantly centrocytes (A, 20X; B, 400X) or centroblasts (D, 20×; E, 400X). The “starry sky” appearance is no longer present. MIB1 indicates a low proliferation rate (C, 20X) in grade 1–2 UFL, and high proliferation rate in grade 3A UFL (F, 20X).


Supplemental Digital Content 4:

Histologic and immunophenotypic features of PNMZL. The nodal architecture is altered with expansion of the interfollicular region by monotonous, small to medium sized cells with monocytoid and plasmacytoid features (A, 40X; B, 400X). CD20 stain shows increased interfollicular B cells; IgD highlights the irregular mantle cuffs (D, 40X).


Supplemental Digital Content 5:

PD-1 staining patterns of normal and neoplastic lymph nodes. PD-1 positive cells in reactive follicles show a polarized distribution pattern towards the light zone (A). Many fewer intrafollicular PD-1 positive cells are seen in the tonsillar (B) and nodal (C) PFL cases. A UFL grade 1–2 case shows markedly increased intrafollicular PD-1 positive cells (D).


This work was supported by funding from the intramural Research Program of the Center for Cancer Research, National Cancer Institute, National Institutes of Health. R.S. is supported by the Kinderkrebsinitative (KKI) Buchholz, Holm-Seppensen.


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The authors have no conflict of interest or other funding to disclose.


1. Harris NL, Swerdlow SH, Jaffe ES, et al. Follicular lymphoma. In: Swerdlow SH, Campo E, Harris NL, et al., editors. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: International Agency for Research on Cancer; 2008. pp. 220–226.
2. Atra A, Meller ST, Stevens RS, et al. Conservative management of follicular non-Hodgkin's lymphoma in childhood. Br J Haematol. 1998;103:220–223. [PubMed]
3. Lorsbach RB, Shay-Seymore D, Moore J, et al. Clinicopathologic analysis of follicular lymphoma occurring in children. Blood. 2002;99:1959–1964. [PubMed]
4. Oschlies I, Salaverria I, Mahn F, et al. Pediatric follicular lymphoma--a clinico-pathological study of a population-based series of patients treated within the Non-Hodgkin's Lymphoma--Berlin-Frankfurt-Munster (NHL-BFM) multicenter trials. Haematologica. 2010;95:253–259. [PubMed]
5. Agrawal R, Wang J. Pediatric follicular lymphoma: a rare clinicopathologic entity. Arch Pathol Lab Med. 2009;133:142–146. [PubMed]
6. Frizzera G, Murphy SB. Follicular (nodular) lymphoma in childhood: a rare clinical-pathological entity. Report of eight cases from four cancer centers. Cancer. 1979;44:2218–2235. [PubMed]
7. Pinto A, Hutchison RE, Grant LH, Trevenen CL, Berard CW. Follicular lymphomas in pediatric patients. Mod Pathol. 1990;3:308–313. [PubMed]
8. Swerdlow SH. Pediatric follicular lymphomas, marginal zone lymphomas, and marginal zone hyperplasia. Am J Clin Pathol. 2004;122(Suppl):S98–S109. [PubMed]
9. Taddesse-Heath L, Pittaluga S, Sorbara L, Bussey M, Raffeld M, Jaffe ES. Marginal zone B-cell lymphoma in children and young adults. The American journal of surgical pathology. 2003;27:522–531. [PubMed]
10. Shao H, Xi L, Raffeld M, et al. Nodal and extranodal plasmacytomas expressing immunoglobulin a: an indolent lymphoproliferative disorder with a low risk of clinical progression. The American journal of surgical pathology. 2010;34:1425–1435. [PMC free article] [PubMed]
11. Masir N, Campbell LJ, Jones M, Mason DY. Pseudonegative BCL2 protein expression in a t(14;18) translocation positive lymphoma cell line: a need for an alternative BCL2 antibody. Pathology. 2010;42:212–216. [PubMed]
12. Segal GH, Jorgensen T, Masih AS, Braylan RC. Optimal primer selection for clonality assessment by polymerase chain reaction analysis: I. Low grade B-cell lymphoproliferative disorders of nonfollicular center cell type. Hum Pathol. 1994;25:1269–1275. [PubMed]
13. Ramasamy I, Brisco M, Morley A. Improved PCR method for detecting monoclonal immunoglobulin heavy chain rearrangement in B cell neoplasms. J Clin Pathol. 1992;45:770–775. [PMC free article] [PubMed]
14. Kononen J, Bubendorf L, Kallioniemi A, et al. Tissue microarrays for high-throughput molecular profiling of tumor specimens. Nat Med. 1998;4:844–847. [PubMed]
15. Salaverria I, Philipp C, Oschlies I, et al. Translocations activating IRF4 identify a subtype of germinal center-derived B-cell lymphoma affecting predominantly children and young adults. Blood. 2011;118:139–147. [PubMed]
16. Salaverria I, Akasaka T, Gesk S, et al. The CBFA2T3/ACSF3 locus is recurrently involved in IGH chromosomal translocation t(14;16)(q32;q24) in pediatric B-cell lymphoma with germinal center phenotype. Genes Chromosomes Cancer. 2012;51:338–343. [PubMed]
17. Kussick SJ, Kalnoski M, Braziel RM, Wood BL. Prominent clonal B-cell populations identified by flow cytometry in histologically reactive lymphoid proliferations. Am J Clin Pathol. 2004;121:464–472. [PubMed]
18. Berget E, Helgeland L, Molven A, Vintermyr OK. Detection of clonality in follicular lymphoma using formalin-fixed, paraffin-embedded tissue samples and BIOMED-2 immunoglobulin primers. Journal of clinical pathology. 2011;64:37–41. [PubMed]
19. Martinez A, Pittaluga S, Rudelius M, et al. Expression of the interferon regulatory factor 8/ICSBP-1 in human reactive lymphoid tissues and B-cell lymphomas: a novel germinal center marker. The American journal of surgical pathology. 2008;32:1190–1200. [PubMed]
20. Sciammas R, Shaffer AL, Schatz JH, Zhao H, Staudt LM, Singh H. Graded expression of interferon regulatory factor-4 coordinates isotype switching with plasma cell differentiation. Immunity. 2006;25:225–236. [PubMed]
21. Falini B, Fizzotti M, Pucciarini A, et al. A monoclonal antibody (MUM1p) detects expression of the MUM1/IRF4 protein in a subset of germinal center B cells, plasma cells, and activated T cells. Blood. 2000;95:2084–2092. [PubMed]
22. Karube K, Guo Y, Suzumiya J, et al. CD10-MUM1+ follicular lymphoma lacks BCL2 gene translocation and shows characteristic biologic and clinical features. Blood. 2007;109:3076–3079. [PubMed]
23. Muenst S, Dirnhofer S, Tzankov A. Distribution of PD-1+ lymphocytes in reactive lymphadenopathies. Pathobiology. 2010;77:24–27. [PubMed]
24. Muenst S, Hoeller S, Willi N, Dirnhofera S, Tzankov A. Diagnostic and prognostic utility of PD-1 in B cell lymphomas. Dis Markers. 2010;29:47–53. [PMC free article] [PubMed]
25. Finn LS, Viswanatha DS, Belasco JB, et al. Primary follicular lymphoma of the testis in childhood. Cancer. 1999;85:1626–1635. [PubMed]
26. Campo E, Swerdlow SH, Harris NL, Pileri S, Stein H, Jaffe ES. The 2008 WHO classification of lymphoid neoplasms and beyond: evolving concepts and practical applications. Blood. 2011;117:5019–5032. [PubMed]
27. Fend F, Cabecadas J, Gaulard P, et al. Early lesions in lymphoid neoplasia: Conclusions based on the Workshop of the XV. meeting of the European Association of Hematopathology and the Society of Hematopathology in Uppsala, Sweden. Journal of Hematopathology. in press. [PMC free article] [PubMed]