The paper, “Different bone mineral density in cervical and endometrial cancer,” came from a group of Soonchunhyang University and was published online late last year. It purported to look at the association between gynecologic cancers and bone mineral density:
Objectives Patients with cervical cancer have lower bone mass than women without cancer, whereas women with endometrial cancer have higher bone mineral density (BMD) than control subjects, possibly due to the prevalence of high body-fat mass. The aim of this study was to compare BMD in patients with cervical cancer, endometrial cancer and controls. Methods We analyzed and compared spinal and femoral BMD in 130 patients with cervical cancer, 68 with endometrial cancer, and 140 age-matched menopausal female control subjects. We also compared serum calcium, phosphorus, total alkaline phosphatase, osteocalcin, and urinary deoxypyridinoline levels. Results Compared with the control group, T-scores for some lumbar vertebrae (L4), the femoral neck, and Ward’s triangle were lower in patients with cervical cancer, whereas only L4 T-scores were significantly lower in patients with endometrial cancer. Deoxypyridinoline levels were significantly lower in women with endometrial cancer (p < 0.002) than in women with cervical cancer, but no other biochemical variables differed among groups. Conclusions Cervical cancer was associated with lower BMD, especially in femoral BMD, and may be a risk factor for secondary osteoporosis. However, endometrial cancer generally seemed to have no damaging effect on bone except at L4. A further larger follow-up study in more populations is required to clarify these findings.
But it didn’t take long before this notice appeared:
The Editors and Publisher regret to announce that the following article published online in 2013 has been retracted at the request and agreement of the first author, on behalf of all authors, from publication in Climacteric.
Lee ES, Park BR, Kim A, Choi CH, Kim HY. Different bone mineral density in cervical and endometrial cancer. Climacteric 2013; Epub November 11. (DOI: 10.3109/13697137.2013.850479).
E. S. Lee writes:
Upon reviewing, similarity was noticed between ‘Bone mineral density of lumbar spine and femur in patients with gynecologic cancer’ by S Chun MD which is accepted manuscript for publication in this journal (Climacteric) and my article.
Both articles are based on data collected from the same clinical practice of HY Kim, MD, who is also appointed as a correspondent of my paper.
I am writing to withdraw the article before it is finally printed in the Journal in order that a potentially overlapping and undesirable publication would be avoided and our professional ethics be upheld.
On behalf of all the co-authors, I sincerely apologize for any inconvenience caused to you.
The article is herewith withdrawn from all print and electronic editions.
What about that Chun paper? Well, it, too, included 130 patients with cervical cancer and 68 women with endometrial cancer (but a different number of controls):
Objective To compare the bone mineral density (BMD) of the lumbar spine and femur in postmenopausal women with cervical and endometrial cancer without bone metastasis with that in normal control postmenopausal women
Methods We retrospectively analyzed the BMD of the lumbar spine and femur using dual-energy X-ray absorptiometry in 130 patients with cervical cancer, 68 patients with endometrial cancer, and 225 healthy controls.
Results The serum levels of calcium, phosphorus, osteocalcin, total alkaline phosphatase and urine deoxypyridinoline were measured in all participants. Age, body mass index, parity and time since menopause were not significantly different between the three groups. The T-scores of basal BMD at the fourth lumbar vertebra (L4) were significantly lower in patients with cervical cancer (− 0.68 ± 0.10) compared to those in the other two groups. Additionally, the incidence of osteoporosis at L4 according to the basal status of bone mass was significantly higher in patients with cervical cancer (10.0%) compared to that in controls (0.4%). Urine deoxypyridinoline levels were significantly higher in patients with cervical cancer compared to those in controls. No differences in basal BMD of the lumbar spine and femur were observed between patients with endometrial cancer and controls, and no significant differences in biochemical markers were detected between patients with endometrial cancer and controls.
Conclusion Our results suggest that postmenopausal women with cervical cancer have a lower BMD and are at increased risk of osteoporosis in the lumbar spine before receiving anticancer treatment compared with postmenopausal women with endometrial cancer.
If we take Lee’s note at face value, it’s still problematic. For starters, the two papers aren’t “potentially overlapping.” They are decidedly overlapping. Second, although there’s no reason to doubt that the data in both articles came from the same source, we don’t see HY Kim’s name on the Chun paper. Does that mean the Chun article failed to adequately cite its own data? Should both articles be pulled?
Then there’s the timing issue. The retracted article appears to have been published a month before the Chun paper. Why was the first one yanked but not the second?
We’ve emailed the journal for comment and will update this post if we learn anything.
Update, 6:15 p.m. Eastern, 2/3/14: Climacteric editor-in-chief Anna Fenton tells us:
I was the editor involved in the manuscript retracted from Climacteric. The data for the two papers was collected from the practice of Dr Kim who is mentioned as a correspondant for the Lee paper. Full access and permission was provided by Dr Kim. I have discussed the issue with all authors and Dr Kim. It was a simple case of mixed communication that lead to 2 manuscripts being written using the same data. There was no plagiarism involved and all authors and Dr Kim are happy with this outcome.