Authors retract highly cited XMRV-prostate cancer link paper from PNAS

Retraction Watch readers may recall that nearly two years ago, an editor at PLOS declared the scientific story of a link between XMRV, aka xenotropic murine leukemia-related virus, and prostate cancer over, saying that a retraction from PLOS Pathogens was the “final chapter.” (That retraction led to an apology from the journal about how it was handled.)

Perhaps, however, there is an epilogue. This week, a group of authors who published a highly cited 2009 study in the Proceedings of the National Academy of Sciences (PNAS) making the same link retracted it. Here’s the notice, signed by all five authors:

Retraction for “XMRV is present in malignant prostatic epithelium and is associated with prostate cancer, especially high-grade tumors,” by Robert Schlaberg, Daniel J. Choe, Kristy R. Brown, Harshwardhan M. Thaker, and Ila R. Singh, which appeared in issue 38, September 22, 2009, of Proc Natl Acad Sci USA (106:16351–16356; first published September 8, 2009; 10.1073/pnas.0906922106).

The authors wish to note, “Due to work performed in other labs, we now know that some conclusions from our paper on xenotropic murine leukemia virus-related virus (XMRV) cannot be true. However, other findings that we reported in that paper still remain valid.

“XMRV was first described in 2006 as a new retrovirus detected in prostate cancer tissues (1). We replicated this finding, identifying XMRV sequences by PCR from an independent set of prostate cancer samples. Other groups also detected XMRV sequences in prostate cancers by PCR (2, 3). However, subsequent studies showed that XMRV was in fact generated by the recombination of two endogenous murine retroviruses when a prostate cancer was passaged in nude mice to generate the 22Rv1 cell line (4). The detection of XMRV DNA in various human tissues by PCR has been attributed to contamination of commercially available reagents with mouse DNA (5).This explanation is the most likely for the PCR findings we reported.

“The immunohistochemical staining with anti-XMRV antiserum that we reported in our PNAS publication was most likely due to cross-reactivity of our antiserum with a protein present almost exclusively in malignant prostatic epithelial cells. We are in the process of identifying this cross-reactive protein.

“We wish to note that other parts of our paper remain valid. We created a full-length infectious clone that replicated efficiently in a human prostate cancer cell line. We used transmission electron microscopy to analyze the XMRV particles produced and showed that their morphology was identical to type-C retroviruses. Using gel electrophoresis and Western blotting, we determined the molecular weights of all the structural and nonstructural proteins of XMRV. Such detailed characterization of a xenotropic virus, including electron microscopy, has not, to our knowledge, been performed elsewhere. This characterization still remains correct and is relevant to the understanding of other wildtype
xenotropic viruses.

“Taking all of this information together, we would like to retract our paper; specifically, the findings reported in Figs. 2–4 and Fig. S1 are no longer valid and we no longer believe that XMRV is associated with prostate cancer.”

The paper has been cited 199 times, according to Thomson Scientific’s Web of Knowledge. The first reference in the retraction is to the retracted 2012 PLOS Pathogens study.

A Science paper claiming a link between XMRV and chronic fatigue syndrome (CFS) was retracted in 2011.