1.

Introduction

Over the past few decades, there has been increasing interest in individual risk assessment for breast cancer.^1–3 Motivations for this include the identification of individuals at extremely high risk who would be potential candidates for risk-reducing surgery or preventive therapy;⁴ delineation of populations at moderately enhanced risk who might benefit from enhanced screening;⁵ and more recently, identification of populations at sufficiently low risk as not to require screening or risk management.⁶

Breast cancer has a relatively well established hormonal aetiology, in addition to a growing body of knowledge on genetic risk factors.^7,8 Although existing risk models have shown a degree of accuracy in prediction [the area under the receiver operating characteristic curve (AUC) ranges around 0.56 to 0.77 for different predictors], it is clear that there is room for improvement.^9,10 One area that offers hope for improved risk assessment is utilization of digital mammographic image features. Mammographic density, which is broadly defined as the amount of radio-opaque tissue, is well known as an important independent risk factor for breast cancer.^11–13 Some previous research has tried to improve mammographic density risk assessment by looking at other image features of a mammogram, and computational advances in machine learning are starting to spur more work.^14–17 A limitation with much of the literature looking at textural or other features from mammograms has been reproducibility.¹⁸

This study was performed to validate a previously developed texture marker as a breast cancer risk feature by testing its use in an independent case–control dataset.¹⁹ The texture marker measures the dispersion of breast density within the mammogram. It was previously found to be associated with breast cancer risk in a case–control study of women in Manchester, UK [odds ratio per standard deviation $(\text{SD}\text{-}\mathrm{OR})=1.36$], where a subgroup analysis suggested it was most predictive for interval cancers that were detected between routine screening rounds ($\text{SD}\text{-}\mathrm{OR}=2.09$).¹⁹ We refer to the marker as MMTEXT for the rest of this paper. Our primary objective was to assess whether MMTEXT is a risk factor for breast cancer in Virginia, USA, after allowing for other classical risk factors from a questionnaire, and for BI-RADS breast density. The prespecified hypothesis was MMTEXT is a risk factor after adjusting for classical risk factors.

2.

Methods

3.

Results

The flow of patients is shown in a flow diagram (Fig. 1). Demographic differences were, as expected, apparent between cases and controls (Table 1): cases were more likely to live further away from UVA than controls and be assessed for financial assistance; controls were more likely than cases to have a higher level of education, private health insurance, and be white. Cases were at a higher risk from classical risk factors (Tyrer-Cuzick risk) after adjustment for age and demographic differences (Table 1). It is interesting to note that BMI has a surprisingly higher $\mathrm{LR}-{\chi }^2$ compared with Tyrer-Cuzick risk (Table 1). However, the effect size of BMI was attenuated in a fully adjusted model (IQ-OR 1.08 for BMI versus 1.53 for Tyrer-Cuzick; Table 3).

Fig. 1

Data flows: sample tested and evaluated for MMTEXT.

Table 1

Demographic and breast cancer risk factors in cases and controls. For the continuous variables the median and interquartile range (IQR) are given in the case and control columns, with OR for the IQR difference in controls. ORs for demographic factors and age are unadjusted; for risk factors other than age they are adjusted for age and the demographic factors shown.

MMTEXT was negatively correlated with age at mammogram (Spearman correlation $\rho =-0.19$) and BMI ($\rho =-0.37$), with an overall nonlinear association shown by Fig. 2. MMTEXT had a small correlation with 10-year risk from the Tyrer-Cuzick model (Spearman $\rho =0.07$, $P=0.007$). MMTEXT was strongly positively associated with BI-RADS breast density (Fig. 3, Spearman $\rho =0.67$).

Fig. 2

Association between MMTEXT and age and BMI in controls. The points show standardized MMTEXT for each woman. The line in the first plot corresponds to the expected MMTEXT for a woman with average BMI, similarly for the second plot. Standard errors are shaded around each line.

Fig. 3

Boxplot distributions of MMTEXT by BIRADS density category.

BI-RADS density was associated with close to a threefold difference in risk between the very dense and fatty categories [OR 2.97 (95%CI 1.58 to 5.57), $\mathrm{LR}-{\chi }_1^2$ 15.2 (trend), aAUC 0.54 (95%CI 0.50 to 0.57)] after adjustment for demographic factors and classical risk factors (Table 2). A much less strong risk difference was observed for MMTEXT [OR 1.27 (95%CI 0.72 to 2.24)] based on frequency matching dense and fatty categories in controls, $\mathrm{LR}-{\chi }_1^2$ 1.7 (continuous), aAUC 0.50 (95%CI 0.49 to 0.54). On a continuous scale for MMTEXT $\mathrm{IQ}\text{-}\mathrm{OR}=1.16$ (95%CI 0.92 to 1.46). There was no predictive information in MMTEXT after further adjustment for BI-RADS density [IQ-OR 0.92 (95%CI 0.76 to 1.10), $P=0.35$]. The fully adjusted logistic regression model fit is shown in Table 3.

Table 2

Comparison between adjusted ORs associated with breast density and MMTEXT in the study. OR adjusted for age, BMI, and demographic and classical risk factors. Continuous odds ratios per interquartile range (IQ-OR) in controls are shown in to four categories with the same number of controls as BI-RADS density, for comparison. *Median (IQR).

Table 3

Multivariable logistic regression analysis of breast cancer risk, showing the impact of adding the imaging biomarker MMTEXT to a base model with variables to adjust based on demographic characteristics, classical risk factors, and mammographic density.

Subgroup analysis by mode of cancer detection suggested potential merit of MMTEXT for cancers that were not detected by mammography [unadjusted for density: IQ-OR 1.46 (0.99 to 2.15) versus 1.03 (0.79 to 1.35), Phet 0.10, aAUC 0.55 (95%CI 0.48 to 0.60); adjusted for density: IQ-OR 1.11 (0.70 to 1.77) versus 0.76 (0.55 to 1.05), Phet 0.21; Table 4].

Table 4

Logistic regression results for breast cancer risk from MMTEXT for subgroups based on mode of detection.

Other exploratory analysis found little evidence for heterogeneity of MMTEXT by age, breast density, or Tyrer-Cuzick risk.

4.

Discussion

This independent study aimed to validate MMTEXT as a risk factor for breast cancer and whether it provides additional information to classical risk factors for risk assessment. Unlike the earlier study,¹⁹ the estimated $\mathrm{IQ}\text{-}\mathrm{OR}=1.16$ (0.92 to 1.46) is, however, not statistically significant at 0.05 level. After further adjustment for density the overall predictive ability of MMTEXT was reduced, indicating that much of the information for risk assessment is related to breast density. There was weak evidence that MMTEXT was more predictive for cancers that were not detected mammography (IQ-OR 1.46, 95%CI 0.99 to 2.15); however, after adjustment for breast density, the predictive ability decreases ($\mathrm{IQ}\text{-}\mathrm{OR}=1.11$, 95%CI 0.70 to 1.77).

This is the first study to assess external validity of MMTEXT and we were able to adjust for classical risk factors using a validated risk model, which is the most important test of a new biomarker.²⁸ Image quality was high because full-field digital mammography was used (not scanned film as much previous work¹⁸) and comparable with the development study.

This study nevertheless differs from the earlier study¹⁹ in a number of aspects, which might partly explain the different findings. The assessment used a different population than the development sample. A notable difference is that all the cancer cases in this study are invasive, and almost a quarter of cases used for variable selection and model training in the development study were ductal carcinoma in situ (DCIS). Invasive cancers differ from DCIS cancers mammographically, as invasive cancers are most often manifest as noncalcified masses so more subtle or occult compared to DCIS cancers.²⁹ It is possible that different composition types of cancers affected the results. The demographic factors between the two studies also differ: for example, the percentage of age over 70 years was around 7% in the development study but around 16% in this validation study; the percentage of non-white was also much higher for cases in this study (17% versus 8%).

Although this study failed to find evidence of predictive ability of MMTEXT, BI-RADS was a strong predictor for breast cancer (also in the wider cohort, see Ref. 30). It is arguable that (currently) there is no better measure of mammographic density than visual assessment from an expert,³¹ and in our previous analysis of mammographic density for risk assessment we found that BI-RADS density conferred slightly more predictive information than a volumetric method on the same data.³⁰ Another cohort study has confirmed the association of BI-RADS density with risk in this case–control study, when used in the combination with classic risk factors during a follow-up of 19 years.²⁰

As discussed above, our subgroup analysis indicate that there may be potential merit of MMTEXT for cancers that were not detected by mammography. If this can be supported by future study with larger sample, there are potential implications for future clinical value. If true, then potentially MMTEXT might have a role as a marker for risk of interval cancer due to masking from mammography. One area this would be useful is to determine eligibility for supplemental screening modalities, such as ultrasound or magnetic resonance imaging. Inspection of the mathematical formula shows that MMTEXT is associated with breast density because it is maximized when white areas of the image are surrounded by other white areas so that images with breast density widely dispersed on the mammogram will have greater MMTEXT values than those with smaller dense areas. A limitation of visual assessment of breast density is the time and expert resources required. In situations where these are important issues, the fully automatic, objective, and freely available MMTEXT might be considered.²¹ MMTEXT currently requires raw mammographic images as it was developed on such images, which may limit its application. Although the same method¹⁹ can also be trained on processed images, making it suitable for processed images, a potential barrier is that manufacturers’ proprietary processing algorithms may result in images less comparable between different machines. It would nevertheless be interesting for a future study to apply the algorithm describe previously¹⁹ and test it on processed images.

There are other limitations of the study include the following. First, controls differed from cases due to geography and other socioeconomic and demographic factors, and we needed to adjust for these differences as far as possible in the analysis. Second, self-reported BMI at the time of questionnaire was used, and we had no validation of the self-reported anthropomorphic measures. This is similar to the development study but is expected to have a minimal impact on the overall findings, because it has been seen elsewhere that self-reported measures are likely to be sufficiently accurate.³² Third, there is a possible survivorship bias because some women died with breast cancer before the questionnaire was available. However, this is unlikely to lead to an overstatement the main findings and is more likely to weaken them, because on average the deceased cases will have been diagnosed at a more advanced stage than those alive and since density is associated with later diagnosis (masking), this bias might be expected to attenuate the predictive ability of MMTEXT. Fourth, it was not possible to include cases who did not respond to the request to complete a questionnaire ($n=47$ aged 40 to 79 years). However, if they had been available then the number of cases would only increase by 10%, and it seems unlikely that nonresponse is associated with mammographic density or MMTEXT other than through the factors adjusted for in the analysis such as age and demographics; this issue is also expected to have minimal impact on the main findings. Finally, both the UVa and earlier Manchester studies were predominantly white women and case–control designs.

In conclusion, data from this study do not support risk assessment for invasive breast cancers using MMTEXT, a fully automatic digital mammographic texture risk factor based on raw (“for processing”) DICOM files. This negative finding demonstrates the importance of external validation. Future studies may focus on developing and validating other measures of risk. MMTEXT may, however, have its potential for cancers not detected primarily due to mammography. Further studies are required to verify this, including longer-term effects in cohort studies.

5.

Appendix

The regression analysis results after adjustment for different machines are shown in Tables 5 and 6 below. The tables include data from women with GE or Hologic machines; data from Hologic machines were excluded in the primary analysis (see Fig. 1).

Table 5

Multivariable logistic regression analysis of breast cancer risk, showing the impact of adding the imaging biomarker MMTEXT to a base model with variables to adjust based on demographic characteristics, classical risk factors, and machine type.

Table 6

Multivariable logistic regression analysis of breast cancer risk, showing the impact of adding the imaging biomarker MMTEXT to a base model with variables to adjust based on demographic characteristics, classical risk factors, machine type, and mammographic density.