Search tips
Search criteria 


Logo of jclinpathJournal of Clinical PathologyVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
J Clin Pathol. 2006 October; 59(10): 1111–1113.
PMCID: PMC1861752

Splitting bone marrow trephines into frozen and fixed fragments allows parallel histological and molecular detection of B cell malignant infiltrates


Clonality analysis of the immunoglobulin heavy chain (IgH) gene is helpful in identifying malignant B cell infiltrates in the bone marrow and is usually carried out on separate aspirates or on the same formalin‐fixed decalcified bone marrow specimen. To determine whether the removal of the decalcification step would improve the molecular analysis, we first studied 12 bone marrow specimens with lymphoma infiltration split into a fixed and a small frozen fragment. Both the detection rate of IgH gene monoclonality and DNA quality were found to be superior in the frozen part than in the fixed part. Conversely, to evaluate whether the split would compromise histological analysis, we selected a series of 134 bone marrow specimens obtained from patients with small B cell lymphoma and showing IgH monoclonality on the frozen part. The histological detection rate of infiltrated or suspicious infiltrates (95%) on the fixed part was not altered by saving a frozen part.

The evaluation of bone marrow biopsy specimens has become a standard procedure in the staging and monitoring of patients with B cell lymphoma. Histological examination easily recognises paratrabecular and diffuse malignant infiltrates, whereas interstitial or focal small B cell lymphoma are difficult to differentiate from benign lymphoid aggregates and may be classified as suspicious for lymphoma.1,2,3

Therefore, polymerase chain reaction (PCR) amplification of the immunoglobulin heavy chain (IgH) gene is increasingly used to assess B cell monoclonality in a wide variety of bone marrow sources, including smears or touch preparations, fresh aspirates, formalin‐fixed paraffin‐wax‐embedded (FFPE) decalcified trephine biopsy specimens and clot sections.1,2,4,5

Reproducible monoclonal IgH rearrangements are highly specific for the presence of malignant B cells in the context of suspected lymphoid infiltrates.2 Conversely, PCR may yield false‐negative results for several reasons. The lack of FR3 region amplification may be due to primer mismatch or somatic hypermutation of IgH.1,4 Moreover, lymphoma cells may be trapped in the bone marrow but absent from the aspirate or smear from which the DNA is extracted.1,4 Combining morphological and molecular analysis on the same FFPE bone marrow fragment would avoid such pitfalls, but fixation or decalcification considerably reduces the scope of amplifying IgH rearrangements.6

We evaluated the performance of molecular testing (DNA quality and FR3‐JH amplification) in a pilot series of 12 bone marrow specimens split into frozen and fixed parts without the decalcification step. Thereafter, we examined histological data in 134 split bone marrow specimens with B cell monoclonality on the frozen part to determine whether splitting had led to false‐negative histological results on FFPE sections.

Materials and methods

The first series included 12 trephine bone marrow biopsy specimens with obvious B‐cell lymphoma infiltration. The frozen part was obtained by cutting a fragment ([less-than-or-eq, slant]4 mm long) of a bone marrow biopsy specimen at least 15 mm long, taken from beneath the cortical bone. The remaining part was fixed in formalin containing 2% acetic acid (48 h) and embedded in paraffin wax without decalcification (Dako, Trappes, France).

Secondly, 134 cases were retrieved from our files (2000–2005) on the basis of the presence of IgH monoclonality on the frozen part in bone marrow specimens. The other part was fixed in Bouin's liquid, decalcified and embedded. Histological analysis included sequential staining and immunohistochemistry for CD20 and CD3 antigens (Dako). Specimens were classified and the results reported as positive, suspect or negative for lymphoma.1,2,3

Finally, a consecutive series of 33 split bone marrow specimens were collected at the initial staging of patients with follicular lymphoma, irrespective of their molecular profile.

Genomic DNA was extracted by a standard phenol–chloroform procedure. DNA quality of 12 bone marrow specimens split into frozen and FFPE fragments was checked in parallel by duplex‐PCR of albumin (200 bp) and ribosomal protein (400 bp) gene amplicons. Clonality analysis of the FR3‐JH IgH gene was carried out as described elsewhere.7


Parallel analysis of frozen and FFPE bone marrow pairs (n = 12)

The histological quality of undecalcified specimens was found to be equivalent to that of Bouin's liquid‐fixed specimens (data not shown). DNA quality control showed amplification of only the 200‐bp amplicons in 7 of 12 (58%) FFPE parts, whereas both 200‐bp and 400‐bp amplicons were detected in all frozen parts (fig 1A1A).). Amplification of monoclonal FR3‐JH IgH amplicons was achieved in 10 of the 12 FFPE parts of DNA, whereas a monoclonal band was detected in all frozen parts (fig 1B1B).). The monoclonal amplicon was always of similar size in FFPE and frozen parts, but its intensity was weaker in three FFPE samples. Two specimens with monoclonal rearrangement on frozen DNA exhibited a false polyclonal profile on FFPE DNA.

figure cp30379.f1
Figure 1 (A) Three examples with parallel amplification of two control gene amplicons in frozen (F) and fixed (f) bone marrow parts. (B) Same examples with parallel amplification of immunoglobulin heavy chain (IgH) gene FR3‐JH rearrangement ...

Histological evaluation of the 134 bone marrow specimens with IgH monoclonality

The 134 bone marrow specimens were obtained from 90 patients either at staging (n = 90) or during follow‐up (n = 44) of patients with mantle (n = 31), follicular (n = 29), lymphocytic (n = 53) or splenic marginal zone (n = 21) B cell lymphoma. The frozen DNA exhibited a reproducible dominant monoclonal FR3‐JH rearrangement. Lymphoma infiltration was seen on fixed sections in 91% of the samples. Histologically, samples suspected of lymphoma (4%) corresponded to bone marrow biopsy carried out either at staging (n = 2) or during follow‐up (n = 3). The negative samples (5%) corresponded to post‐therapeutic samples of seven patients with either follicular or mantle cell lymphoma, followed by recurrence in two patients (table 11).

Table thumbnail
Table 1 Comparison of morphological and molecular analysis on separate bone marrow parts

Splitting bone marrow trephines at staging in 33 patients with follicular lymphoma

In all, 27 (82%) patients exhibited nodular and paratrabecular lymphoma infiltration. Five patients showed both negative histological results and IgH gene polyclonality. Only one patient had a negative histological result, IgH monoclonality in the frozen part.


It is unusual to analyse frozen bone marrow tissue, as saving a frozen fragment could theoretically lead to false‐negative histological results for focal or nodular infiltrates. Biases, such as failing to aspirate sufficient lymphoma cells owing to their entrapment or patchy distribution in the bone marrow, may be avoided by combining histological and molecular testing on the same FFPE sample.1 Unreproducible bands may be observed by using several molecular protocols, including semi‐nested PCR, which is a sensitive technique developed to improve the detection of monoclonal B cells in FFPE specimens.1,2,3,4,8 Both the performance and the interlaboratory reproducibility of IgH gene analysis are poor on FFPE tissues compared with fresh or frozen tissue.6,9 Our study extends these findings to FFPE bone marrow specimens, even without the decalcification step. Such fixation greatly improved the detection of several antigens such as cyclin D1, CD10 and MIB‐1‐Ki‐67 (our personal results). However, only the frozen part proved to be a sufficiently reliable source of optimal DNA to allow the amplification of larger fragments used to detect FR1‐JH rearrangement or translocation genomic breakpoints.10 In comparison with other series,1,2,3 the saving of a small frozen part did not distort our rates of histological detection on the FFPE part of either malignant (91%) or suspicious (4%) lymphoid infiltrates. This was especially checked in a consecutive series of 33 follicular lymphomas, a lymphoma subtype with focal bone marrow infiltration. Finally, splitting bone marrow trephines permits optimal parallel histological and molecular examination, provided the biopsy is of sufficient length.


FFPE - formalin‐fixed paraffin‐wax‐embedded

IgH - immunoglobulin heavy chain

PCR - polymerase chain reaction


Funding: This work was supported by the “Programme Hospitalier pour la Recherche Clinique, CHU Bordeaux”.

Competing interests: None declared.


1. Coad J E, Olson D J, Christensen D R. et al Correlation of PCR‐detected clonal gene rearrangements with bone marrow morphology in patients with B‐lineage lymphomas. Am J Surg Pathol 1997. 211047–1056.1056 [PubMed]
2. Brinckmann R, Kaufmann O, Reinartz B. et al Specificity of PCR‐based clonality analysis of immunoglobulin heavy chain gene rearrangements for the detection of bone marrow involvement by low‐grade B‐cell lymphomas. J Pathol 2000. 19055–60.60 [PubMed]
3. Braunschweig R, Baur A S, Delacretaz F. et al Contribution of IgH‐PCR to the evaluation of B‐cell lymphoma involvement in paraffin‐embedded bone marrow biopsy specimens. Am J Clin Pathol 2003. 119634–642.642 [PubMed]
4. Ben‐Ezra J, Hazelgrove K, Ferreira‐Gonzalez A. et al Can polymerase chain reaction help distinguish benign from malignant lymphoid aggregates in bone marrow aspirates? Arch Pathol Lab Med 2000. 124511–515.515 [PubMed]
5. Kang Y H, Park C J, Seo E J. et al Polymerase chain reaction‐based diagnosis of bone marrow involvement in 170 cases of non‐Hodgkin lymphoma. Cancer 2002. 943073–3082.3082 [PubMed]
6. Bagg A, Braziel R M, Arber D A. et al Immunoglobulin heavy chain gene analysis in lymphomas: a multi‐center study demonstrating the heterogeneity of performance of polymerase chain reaction assays. J Mol Diagn 2002. 481–89.89 [PubMed]
7. de Mascarel A, Dubus P, Belleannee G. et al Low prevalence of monoclonal B cells in Helicobacter pylori gastritis patients with duodenal ulcer. Hum Pathol 1998. 29784–790.790 [PubMed]
8. Fend F, Gschwendtner A, Gredler E. et al Detection of monoclonal B‐cell populations in decalcified, plastic‐embedded bone marrow biopsies with the polymerase chain reaction. Am J Clin Pathol 1994. 102850–855.855 [PubMed]
9. Hoeve M A, Krol A D, Philippo K. et al Limitations of clonality analysis of B cell proliferations using CDR3 polymerase chain reaction. Mol Pathol 2000. 53194–200.200 [PMC free article] [PubMed]
10. van Dongen J J, Langerak A W, Bruggemann M. et al Design and standardization of PCR primers and protocols for detection of clonal immunoglobulin and T‐cell receptor gene recombinations in suspect lymphoproliferations: report of the BIOMED‐2 Concerted Action BMH4‐CT98‐3936. Leukemia 2003. 172257–2317.2317 [PubMed]

Articles from Journal of Clinical Pathology are provided here courtesy of BMJ Publishing Group