Analysis of Circulating Tumor DNA to Monitor Metastatic Breast Cancer
2013 · paper · Biomedicine
Compares tumor DNA circulating in blood with CA 15-3 and circulating tumor cells for tracking metastatic breast cancer in 30 women on therapy. Tumor DNA was detected in 29 of 30 women, tracked tumor burden more closely, and gave the earliest sign of response in 10 of 19.
1 finding
Reference or label Fig. 3A and 3B scatter plots are annotated "r²=0.36" and "r²=0.61", but the legend states "The Spearman correlation coefficient (r) ... was 0.36" and "... was 0.61"; the same values are labelled as both r and r².
- p. 7 Figure 3A and 3B scatter-plot annotations
r2=0.36
- p. 7 Figure 3 caption
The Spearman correlation coefficient (r) between CA 15-3 levels and ctDNA levels across all time points was 0.36 (P<0.001)
- p. 7 Figure 3 panels A and B vs legend
The Spearman correlation coefficient (r) between CA 15-3 levels and ctDNA levels across all time points was 0.36
- p. 7 Figure 3B plot annotation
r 2=0.61
Only the authors can say what is right here.
Judge's note: The page image of p. 7 clearly shows a superscript 2 (r²) in both panel annotations. The legend calls the same numbers the Spearman r. A coefficient and its square cannot both equal 0.36, so either the plot label or the legend is wrong. Only the authors can say which statistic was reported.
Second check: another model agreed
Report a problem with this finding
2 minor (typos and the like; most checked by the judge only)
Wording Abstract: "Among the measures tested, circulating tumor DNA provided the earliest measure of treatment response in 10 of 19 women (53%)", but the only supporting results text (p. 8) says ctDNA "increased at one or more consecutive time points, on average 5 months (range, 2 to 9) before the establishment of progressive disease by means of imaging", i.e., a comparison with CT for progression only.
- p. 1 Abstract, Results
Among the measures tested, circulating tumor DNA provided the earliest measure of treatment response in 10 of 19 women (53%).
- p. 8 Results, CT and Circulating Biomarkers, top of page
In 10 of the 19 patients (53%), levels of circulating tumor DNA increased at one or more consecutive time points, on average 5 months (range, 2 to 9)
- p. 9 Discussion, first paragraph
DNA often provides the earliest measure of treatment response
Only the authors can say what is right here.
Judge's note: Checked the abstract on p. 1 and the results on pp. 7-8. The main text reports 10/19 (53%) only as ctDNA rises occurring before CT-defined progression. CTC increases (7/19) and CA 15-3 increases (9/18) are given separately, but the text does not compare their timing with ctDNA patient by patient. So the abstract and discussion wording ('earliest ... among the measures tested', 'treatment response') claims more than the results sentence shows. Fig. S5, which may hold the comparison, is not among the files. The percentage itself is consistent.
Second check: another model agreed
Report a problem with this finding
Wording Text gives the median ctDNA as "150 amplifiable copies per milliliter of plasma (interquartile range, 9 to 720)", while Fig. 4E labels the quartiles "25% quant. ctDNA (10.35)", "50% quant. ctDNA (150.1)", "75% quant. ctDNA (710.11)".
- p. 4 Results, Quantification of Circulating Tumor DNA in Plasma
150 amplifiable copies per milli-liter of plasma (interquartile range, 9 to 720)
- p. 8 Figure 4E legend labels
25% quant. ctDNA (10.35)
- p. 8 Figure 4E legend labels
75% quant. ctDNA (710.11)
Only the authors can say what is right here.
Judge's note: Viewed the p. 4 text and the Fig. 4E labels on p. 8. The medians agree (150 vs 150.1), but neither quartile in the figure rounds to the IQR bounds in the text (10.35 vs 9; 710.11 vs 720). A different quantile algorithm or analysis set (the Cox models use covariate values at the first visit) might explain the gap, but the paper does not say so. Only the authors can tell which values are intended.
Second check: another model agreed