WSU-WM and BCWM.1 should not be assumed to represent Waldenström macroglobulinemia cell lines

To the editor:

Cell line models of malignancy have been invaluable tools in understanding the genetics and cell biology of cancer. Unfortunately, valid cell lines are lacking for Waldenström macroglobulinemia (WM).

It should not be taken for granted that a cell line reported to have been established from a human tumor represents the malignant clone. Most commonly, putative tumor-derived B-cell lines turn out to be bystander B cells immortalized by spontaneous infection with Epstein-Barr virus (EBV), which has not been reported to transform WM cells. Another problem arises from inadvertent cross-contamination of cell lines. This can be difficult for individual investigators to identify, but the application of DNA fingerprinting techniques at large repositories can allow for precise and unique identification for each cell line. At the German Collection of Microorganisms and Cell Cultures, DSMZ, 29% of human tumor cell lines deposited were recently demonstrated to be false, cross-contaminating cell lines.¹  For example, WSU-ALCL was identified to be a T-cell acute lymphoblastic leukemia (ALL) cell line (CCRF-CEM), and WSU-CLL a pre-B cell ALL cell line (REH).² 

Blood published a report of a putative Waldenström cell line, WSU-WM, in 1993.³  Although the cell line is EBV-negative, no evidence is presented that it is derived from the index patient. Yet evidence is presented that it is unrelated to the malignant clone: the patient's WM expressed IgM-κ, whereas the cell line expresses IgM-λ. The authors suggest that this could be the result of switching of the light chain, but, given the concerns listed above, this would appear to be the least likely explanation. Although not widely used for many years until a report in this journal in 2003,⁴  it has been used extensively since then.^5-14  Until proven otherwise, the WSU-WM cell line should not be viewed as representing a genuine cell line established from the malignant WM clone in this patient.

In 2007 a report appeared in Experimental Hematology describing another putative WM cell line, BCWM.1,¹⁵  use of which has also been reported in several recent publica-tions.^9-14,16,17  This study does not report the light chain secreted by the primary tumor. The study's authors performed single-nucleotide polymorphism analysis on the cell line and tumor but do not present the data to indicate that the cell line is derived from the index patient. They performed gene expression profiling on the cell lines and tumor, but this does not confirm a clonal relationship. Regrettably, the one test that could have confirmed or refuted a clonal relationship, IgH CDR3 length analysis, was reported for the cell line but not for the tumor. Finally, the authors reported that the cell line expresses EBV latent membrane protein 1 (LMP1). Until proven otherwise, this cell line should be assumed to be a lymphoblastoid cell line that was derived by EBV transformation of a bystander B cell.

The problems described here are by no means unique to WM or to any particular set of investigators. Despite calls to the contrary, EBV-transformed B-cell lines derived from multiple myeloma (MM) patients continue to be used as models of MM, including ARK, ARH77, MC/CAR, HS-Sultan, and UCLA-1. Furthermore, the extent of cross-contamination of cell lines occurring within labs that carry multiple different lines is often unknown or overlooked.

Given the powerful molecular tools that can be used to verify the identity of established lines, it is important to define a unique set of genetic markers for each line so that individual labs can readily confirm that cell line mix-ups are not a complicating issue for their studies.