The Promise of Metabolomics to Transform Breast Cancer Screening

EFFORTS TO ENHANCE breast cancer screening continue, with methods to improve prognostic capabilities undergoing evaluation. One new frontier of cancer detection is emerging: metabolomics.¹ This approach, which quantifies metabolic signatures, has the potential to change how breast cancer is detected, offering greater reliability and less discomfort for the patient. Findings from recent studies in metabolomics show promising capabilities to accurately identify cancer markers using metabolic signatures.^2,3

In the second installment of this 3-part series, the focus is on advancements in metabolomics specifically for breast cancer. Other articles in this series explore metabolomics in gynecological and pancreatic cancers, examining the implications for future oncology treatments.

Study 2 Evaluating Plasma Ion Intensity In this next study, investigators evaluated plasma ion intensity differences in 55 patients with breast cancer and 55 healthy controls, identifying 792 ions with significant changes.3 They evaluated plasma samples using untargeted ultrahigh-performance liquid chromatography paired with mass spectrometry (LC-MS). They then used generalized least absolute shrinkage and selection operator (LASSO) to narrow down to 16 ions—12 in positive mode and 4 in negative mode—that demonstrated strong diagnostic potential and achieved an AUC value of 0.9729 (95% CI, 0.96-0.98). Generalized LASSO is a statistical method used for variable selection and regularization in regression models. Differences were found in amino acid derivatives, such as increased levels of N-stearoyl tryptophan and decreased levels of uracil derivative, along with changes in fatty acid biosynthesis and lipid metabolism–related compounds, including a decrease in caproleic acid. Patients included in the study had an average age of 53 years (± 12.26 years). In terms of menopausal status, 32.7% were premenopausal, 3.64% were perimenopausal, and 63.6% were postmenopausal. Participants were predominantly White (60%), followed by African American (27%), Asian (9%), and Hispanic (3%) participants. Regarding smoking history, 12.7% were current smokers, 9% were former smokers, and 78.1% had never smoked. Histological types of breast cancer included 18.1% with ductal carcinoma in situ, 63.6% with invasive ductal carcinoma, 9% with invasive lobular carcinoma, and 3.6% with mixed types. The cancer stages among the patients were as follows: stage 0 (16.3%), stage I (38.1%), stage II (27.2%), and stage III (7.2%). The grades of breast cancer were categorized as low (12.7%), intermediate (38.1%), and high (45.4%). Lymph node involvement was noted in 25.4% of patients, and 72.7% had no involvement. In terms of hormone receptor status, 81.8% had positive test results for estrogen receptors and 60% had positive test results for progesterone receptors. HER2 status was positive in 20% of patients and negative in 70%. Surgical procedures varied, with 32.7% undergoing bilateral mastectomy, 25.4% having preventive mastectomy, and 40% undergoing mastectomy. Only 1.82% had endoscopy-assisted breast surgery. In addition, 45.4% of patients had palpable tumors and 54.5% did not. Percentages may not total 100% due to missing data, investigators noted. “Our findings are consistent with the findings of other reported studies, especially in the case of amino acid derivatives and fatty acid precursors. Variations in findings across studies could possibly arise due to differences in study designs: sample specimen used, metabolite extraction protocols, LC-MS methodology, and choice of statistics in addition to previous treatment received and other exogenous factors,” study investigators stated.

Study 1 Evaluating Plasma

In the exploratory set of a study by da Silva et al, investigators compared plasma samples of 59 women with untreated stage III invasive breast cancer with 31 samples from women who were at low risk for breast cancer.² The investigators identified differences in key metabolites using targeted quantitative mass spectrometry along with unsupervised clustering analysis.

They found that women in the study with breast cancer had at least 1 metabolite that was either higher or lower in 5 of 7 main metabolite groups measured in their plasma. Glutamine levels in these patients were reduced to approximately one-eighth of the levels found in healthy individuals (approximately 800 μM/L), with a P value of 7.8e to 53. In contrast, aspartate and glutamate levels were almost 10 times higher than normal levels, with P values of 1.7e to 67 and 6.4e to 96, respectively. The area under the curve (AUC) was 0.987 (95% CI, 0.964-1.000).

The reason for this is because the body stores energy sources such as glutamine, which can be characterized in cancer metabolism. For instance, it is the glutamate to glutamine ratio that can diagnose breast cancer with nearly 90% accuracy, Robert A. Nagourney, MD, lead study author, explained. Nagourney is medical and laboratory director at Rational Therapeutics, Inc, in Long Beach, California, and teaches pharmacology at the University of California, Irvine.

“With a random blood sample [result] that showed an abnormal glutamate to glutamine ratio, I would be concerned that the patient had breast cancer. Imagine the power of something like that,” he told Targeted Therapies in Oncology in an interview.

The decrease in glutamine and the increase in aspartate and glutamate suggest a process called glutaminolysis, which is often linked to cancer growth driven by the MYC gene. This led investigators to explore other related MYC-driven changes to better understand the findings. In the validation set comprising 832 samples in the control vs 154 with breast cancer, the AUC was 0.995 (95% CI, 0.991-0.998).

Targeted Metabolomic Analysis

Investigators used a method called targeted metabolomic analysis to examine plasma and tissue samples with a tool by biocrates called AbsoluteIDQ p180. This tool is designed to quickly and accurately measure various metabolites, and investigators were able to precisely measure the levels of 186 specific metabolites in blood using a technique called electrospray ionization tandem mass spectrometry. This analysis was conducted on 1302 samples blinded to any phenotype information, according to the investigators.

Study 2 Evaluating Plasma Ion Intensity

In this next study, investigators evaluated plasma ion intensity differences in 55 patients with breast cancer and 55 healthy controls, identifying 792 ions with significant changes.³ They evaluated plasma samples using untargeted ultrahigh-performance liquid chromatography paired with mass spectrometry (LC-MS). They then used generalized least absolute shrinkage and selection operator (LASSO) to narrow down to 16 ions—12 in positive mode and 4 in negative mode—that demonstrated strong diagnostic potential and achieved an AUC value of 0.9729 (95% CI, 0.96-0.98). Generalized LASSO is a statistical method used for variable selection and regularization in regression models. Differences were found in amino acid derivatives, such as increased levels of N-stearoyl tryptophan and decreased levels of uracil derivative, along with changes in fatty acid biosynthesis and lipid metabolism–related compounds, including a decrease in caproleic acid.

Patients included in the study had an average age of 53 years (± 12.26 years). In terms of menopausal status, 32.7% were premenopausal, 3.64% were perimenopausal, and 63.6% were postmenopausal. Participants were predominantly White (60%), followed by African American (27%), Asian (9%), and Hispanic (3%) participants.

Regarding smoking history, 12.7% were current smokers, 9% were former smokers, and 78.1% had never smoked. Histological types of breast cancer included 18.1% with ductal carcinoma in situ, 63.6% with invasive ductal carcinoma, 9% with invasive lobular carcinoma, and 3.6% with mixed types.

The cancer stages among the patients were as follows: stage 0 (16.3%), stage I (38.1%), stage II (27.2%), and stage III (7.2%). The grades of breast cancer were categorized as low (12.7%), intermediate (38.1%), and high (45.4%). Lymph node involvement was noted in 25.4% of patients, and 72.7% had no involvement.

In terms of hormone receptor status, 81.8% had positive test results for estrogen receptors and 60% had positive test results for progesterone receptors. HER2 status was positive in 20% of patients and negative in 70%.

Surgical procedures varied, with 32.7% undergoing bilateral mastectomy, 25.4% having preventive mastectomy, and 40% undergoing mastectomy. Only 1.82% had endoscopy-assisted breast surgery. In addition, 45.4% of patients had palpable tumors and 54.5% did not. Percentages may not total 100% due to missing data, investigators noted.

“Our findings are consistent with the findings of other reported studies, especially in the case of amino acid derivatives and fatty acid precursors. Variations in findings across studies could possibly arise due to differences in study designs: sample specimen used, metabolite extraction protocols, LC-MS methodology, and choice of statistics in addition to previous treatment received and other exogenous factors,” study investigators stated.

Current Screening Issues

Current breast cancer detection methods, such as mammograms, risk eroding public trust in their effectiveness.¹ Results from a recent study involving 1,053,672 women aged 40 to 73 years without a breast cancer diagnosis found that between 2005 and 2017, mammograms identified 3,184,482 true-negative results but also produced 345,343 false-positive results.

Investigators determined that women were more likely to return for follow-up after receiving a true-negative result (76.9%; 95% CI, 75.1%-78.6%) vs those recalled after a false-positive result for additional imaging (an adjusted absolute difference of –1.9 percentage points; 95% CI, –3.1 to –0.7), short-interval follow-up (–15.9 percentage points; 95% CI, –19.7 to –12.0), or biopsy (–10.0 percentage points; 95% CI, –14.2 to –5.9).

False Positives

Asian and Hispanic women experienced the largest declines in likelihood of returning after a false-positive result, especially when recommended for short-interval follow-up (–20 to –25 percentage points) or biopsy (–13 to –14 percentage points) vs true-negative results. Among women who had 2 screening mammograms within 5 years, a false-positive result on the second examination reduced the likelihood of returning for a third, regardless of the outcome of the first screening.

“Women were less likely to return to screening after false-positive mammography results, especially with recommendations for short-interval follow-up or biopsy, raising concerns about continued participation in routine screening among these women at increased breast cancer risk,” investigators stated. Although the data on categorizing metabolic signatures for breast cancer detection offer promise, the process is still early, a 2023 review explains.⁴

“Despite its success in applications to epidemiology, studies of larger sample size with detailed information on menopausal status, breast cancer subtypes, and repeated biologic samples collected over time are needed to improve comparison of results between studies and enhance validation of results, allowing potential clinical translation of findings,” His et al stated.4 However, if developed, a simple blood draw to detect breast cancer could become a practical and actionable alternative to current methods, depending on the validation of results.

REFERENCES

1. Miglioretti DL, Abraham L, Sprague BL, et al. Association Between False-Positive Results and Return to Screening Mammography in the Breast Cancer Surveillance Consortium Cohort. Ann Intern Med. 2024;177(10):1297-1307. doi:10.7326/ M24-0123

2. da Silva I, da Costa Vieira R, Stella C, et al. Inborn-like errors of metabolism are determinants of breast cancer risk, clinical response and survival: a study of human biochemical individuality. Oncotarget. 2018;9(60):31664-31681. doi:10.18632/ oncotarget.25839

3. Da Cunha PA, Nitusca D, Canto LMD, et al. Metabolomic analysis of plasma from breast cancer patients using ultra-high-performance liquid chromatography coupled with mass spectrometry: an untargeted study. Metabolites. 2022;12(5):447. doi:10.3390/metabo12050447

4. His M, Gunter MJ, Keski-Rahkonen P, Rinaldi S. Application of metabolomics to epidemiologic studies of breast cancer: new perspectives for etiology and prevention. J Clin Oncol. 2024;42(1):103-115. doi:10.1200/jco.22.02754