Sage Journals: Discover world-class research

Abstract

Despite initial treatment, breast cancer recurrence affects approximately 30% of patients. Currently, there exists no standardized approach to detect recurrence before clinical or radiologic signs manifest. Circulating tumor DNA (ctDNA) is a minimally invasive blood test that offers potential to monitor molecular disease and individualize care. This study explores the utility of ctDNA in recurrence monitoring and clinical decision-making for high-risk breast cancer cases within a community setting. Seventy-two patients with high-risk breast cancer features—defined as stage III disease, triple-negative or HR−/HER2+ following neoadjuvant treatment, metastatic breast cancer without evidence of disease, bilateral breast cancer history, high-risk genetics (BRCA1/BRCA2 mutations), <40 years old at diagnosis, or history of breast cancer recurrence—were offered tumor-informed ctDNA assays at 3- to 6-month intervals. Analysis was conducted on 67 cases with a mean diagnostic age of 52.69 at diagnosis. The cohort was ethnically diverse, including White (n = 21, 31.82%), Japanese (n = 15, 22.73%), Native Hawaiian (n = 11, 16.67%), and Filipino (n = 7, 10.61%) patients. Seven (10.45%) tests were positive: six predicted recurrence despite four with initially negative radiological findings, and one prompted treatment resumption following prior non-adherence. However, one negative result was false and later showed a contralateral breast recurrence, and another negative test coincided with a new primary cholangiocarcinoma. In two cases, ctDNA negativity was utilized to monitor treatment response in metastatic disease and inform therapeutic adjustments. In real-world settings, ctDNA served as a valuable tool for earlier recurrence prediction, expediting radiological confirmation, and influencing treatment. Nevertheless, false results carry the risk of hindering effective care and inducing considerable patient anxiety. Future large-scale studies are warranted in high-risk breast cancer populations to evaluate ctDNA’s impact on patient survival outcomes, treatment monitoring, and patients’ emotional experiences.

Plain language summary

How ctDNA testing has been used by physicians managing high-risk breast cancer patients in community clinics

Why was this study done? Breast cancer can come back after initial treatment in about 30% of patients. Currently, there is no standard way to detect these recurrences early, before symptoms or imaging show signs of cancer. A blood test called circulating tumor DNA (ctDNA) may help monitor cancer activity and guide care more effectively. This study explored how ctDNA testing might help in high-risk breast cancer cases in community healthcare settings.

What did the researchers do? The study included 72 patients with high-risk breast cancer. These patients had factors like advanced-stage cancer, aggressive cancer types, genetic risks, or a history of recurrence. Blood samples were collected every 3 to 6 months for ctDNA testing.

What did the researchers find? Out of 67 cases analyzed, ctDNA tests were positive in seven patients. Six of these positive tests correctly predicted cancer recurrence before imaging could confirm it. One positive test occurred after a patient resumed treatment following a gap in care. Two patients used ctDNA testing to monitor their treatment response for metastatic breast cancer. However, there were also challenges. One negative test result was inaccurate, missing a breast cancer recurrence in the opposite breast.

What do these findings mean? This study showed that ctDNA testing could help predict cancer recurrence earlier, guide treatment choices, and improve monitoring for some patients. However, inaccurate results can lead to missed diagnoses and patient anxiety. More research is needed to understand how this testing affects patient survival, care decisions, and emotional well-being.

Keywords

breast cancer recurrence circulating tumor DNA case series high-risk breast cancer patients minimal residual disease (MRD) case series

Background/Introduction

Breast cancer is a globally prevalent cancer diagnosis, ranking second in cancer-related deaths among women in the United States.¹ Despite initial treatment success, up to 30% of patients with breast cancer will experience recurrence.^2,3 Recurrence often is highest within the first 5 years after diagnosis, either returning locally at the site of the original tumor, regionally into neighboring lymph nodes, or distantly into tissues and organs beyond the initial site as metastasis.⁴ About 20% of patients develop recurrence at a later stage,⁵ and if metastatic, suffer a poor prognosis with a 5-year survival of about 68%.⁶ Risk factors for recurrence vary by subtype but include lymph node involvement, large tumor size, younger age (<40) at the time of diagnosis, and obesity.⁷

To identify possible recurrence early and improve overall survival, current National Comprehensive Cancer Network (NCCN) surveillance protocols suggest regular clinical exams 1–4 times per year, continued annual breast imaging with mammography, breast MRI when applicable, and lifestyle education. Screening for metastasis with laboratory or imaging studies is not recommended, and these are indicated only if clinical signs and symptoms are suggestive of recurrent disease.⁸ No standardized tool exists to test for early recurrence risk before clinical signs manifest and radiologic tests confirm suspicious clinical symptoms.

Circulating tumor DNA (ctDNA) has emerged as a promising biomarker for residual disease monitoring and predicting breast cancer recurrence trends. Various commercial platforms are available, including both tumor-informed and tumor-agnostic ctDNA, for current use in patients with high-risk breast cancer. ctDNA assays utilize liquid biopsies to detect tumor-specific mutations that may indicate minimal residual disease (MRD) and impending recurrence.^9–11 Tumor-informed ctDNA tests sequence apoptotic tumor DNA in the bloodstream for somatic variants through whole-exome sequencing of the primary tumor sample. Early mutations that occur at the tumor origin are identified to create a patient-specific panel that can detect early mutations.

The benefits of tumor-informed ctDNA include its minimally invasive approach to molecular tumor profiling and a 90% sensitivity and reproducibility of mutated allele detection.¹² Since ctDNA has a short half-life of up to 150 min, it can effectively monitor a tumor’s response to therapy in real time and consequently enhance treatment efficacy standards.¹³ Clinical applications of ctDNA testing have been described, including the identification of tumor mutations, screening for the early detection of cancer, tracking of therapeutic responses, predicting MRD, and monitoring patients for recurrence.^9–11 In studies assessing breast cancer recurrence, ctDNA positivity has been reported to demonstrate high sensitivity and specificity,¹⁴ with median lead times from clinical relapse ranging from 8.9¹⁵ to 10.7 months, even across all four subtypes.¹⁶

Studies involving patients with high-risk breast cancer emphasize ctDNA’s negative prognostic capabilities and the importance of frequent surveillance to optimize lead time. The I-SPY2 trial assessing ctDNA following neoadjuvant chemotherapy (NACT) in patients with high-risk early-stage luminal A breast cancer and patients with Triple Negative Breast Cancer (TNBC) found that the persistent presence of ctDNA at any time point of NACT (before, during, and after) predicted poorer response and metastatic recurrence, regardless of pathological complete response (pCR).¹⁷ They also found that early clearance of ctDNA significantly predicted pCR, implying its utility as a surrogate measure of tumor burden comparable to radiological imaging, particularly in aggressive subtypes like TNBC. Results from the c-TRAK TN trial, which prospectively tracked molecular residual disease and pembrolizumab activity in 161 patients with moderate-to-high-risk early-stage TNBC, further theorized that frequent testing at less than 3-month intervals may allow earlier interventions to expedite ctDNA clearance, especially given its median 10.7-month lead time.¹⁸

The phase III monarchE trial, which demonstrated the efficacy of adjuvant abemaciclib (CDK4/6 inhibitor) targeted therapy with endocrine therapy (ET) in patients with HR+/HER2+ node-positive high-risk early-stage breast cancer, also assessed the prognostic capabilities of ctDNA (Signatera, Inc., Austin, TX, USA). This trial affirmed the negative prognostic value of ctDNA, with patients who persistently tested positive tending to develop more distant recurrence and longer lead times (median of 15 months). In addition, results contended that ctDNA’s potential to monitor treatment response, as ctDNA positivity mirrored favorable treatment responses in concordance with the original trial findings, with 40% of participants initially ctDNA positive becoming undetectable and 42% experiencing an invasive disease-free survival event following ET + abemaciclib treatment upgrading.¹⁹

Overall, studies have demonstrated the potential of ctDNA across various subtypes, with lead times as low as 9–11 months, offering insights into its clinical implications for prognosis and monitoring treatment response. Although multiple uses of ctDNA have been speculated, Medicare and Medicaid Services approved coverage reimbursement for the tumor-informed Signatera assay strictly for adjuvant and recurrence monitoring in stage IIB and III breast cancers.²⁰

With few studies investigating molecular recurrence in patients with high-risk breast cancer, our case series aims to assess how ctDNA is currently being used in clinical practice, with a focus on clinician decision-making and outcomes, incorporating insights from breast cancer populations in a community setting in Hawaii, where breast cancer incidence rates remain the highest in the nation.²¹

Methods: Patient information, diagnostic assessment, and therapeutic intervention

We conducted a case series of 72 patients with high-risk breast cancer managed at a single outpatient center under the Signatera early access program to investigate how ctDNA is utilized in the clinic. High-risk status was defined by the following criteria: stage III disease, TNBC, or HR−/HER2+ breast cancer following neoadjuvant treatment, metastatic breast cancer without evidence of disease, bilateral breast cancer history, high-risk genetics (i.e., BRCA1 or BRCA2 mutations), young age at diagnosis (<40 years old), or a history of breast cancer recurrence of any subtype.

Patients eligible for the study were prospectively enrolled on a rolling basis from March 2021 to August 2024. All participants underwent personalized tumor-informed ctDNA assays (Signatera™; Natera, Inc., Austin, TX, USA), which involved whole-exome sequencing of the primary tumor and matched normal blood samples to identify somatic clonal variants. For each patient, a personalized panel of 16 tumor-specific variants was selected based on clonality, detectability, and variant allele frequency. This bespoke panel served as a molecular fingerprint to monitor for molecular recurrence throughout the patient’s disease course. Plasma cell-free DNA was isolated, PCR-amplified, and sequenced, with a lower limit of 0.01% variant allele frequency.

Blood samples were taken at baseline and 3-to-6-month intervals for up to 2 years. Those with detectable positive ctDNA results were subsequently offered diagnostic imaging for further evaluation. Through the early access program, participants with undetectable negative ctDNA results had the option to continue testing for up to 5 years from their initial test, with the same option extended to patients with positive ctDNA results. Participants could choose to discontinue testing at any time for any reason.

Demographic data (age at time of diagnosis, reported race and ethnicity, gender), tumor clinical/pathological characteristics (tumor size, grade, node status, hormone receptor status, recurrence, metastasis), treatment history (surgery, chemotherapy, radiation, ET), pCR, tumor-infiltrating lymphocytes, and biomarker test results (genetic testing, genomic testing, and liquid biopsy) were recorded for each patient. Particular attention was given to illustrative or clinically impactful cases, with a focus on the predictive value of ctDNA as well as its influence on treatment decisions made by clinicians.

The reporting of this study conforms to the Consensus-based Clinical CAse REporting (CARE) Guideline statement²² (CARE-checklist).

Results: Clinical findings, follow-up, and outcomes

A total of 72 patients registered between March 2021 and August 2024 were offered ctDNA tumor sequencing. Of these, five patients did not obtain ctDNA testing and were excluded from further analysis, leaving 67 patients for final review. The age of patients ranged from 28 to 83 years with an average age of 52.69 years old. The cohort was racially and ethnically diverse: White (n = 21, 31.82%), Vietnamese (n = 1, 1.52%), Korean (n = 2, 3.03%), Japanese (n = 15, 22.73%), Chinese (n = 5, 7.58%), Filipino (n = 7, 10.61%), Native Hawaiian (n = 11, 16.67%), Black (n = 2, 3.03%), and Hispanic (n = 3, 4.55%).

An average of four ctDNA tests were conducted per patient over 2 years, ranging from 1 to 10 tests. Fifty-five patients completed at least one ctDNA test. Seven (10.45%) tests were positive: three predicted metastatic diagnosis despite initially negative radiological findings (a52), with two prompting early treatment escalation (a05, a57); one predicted local recurrence following treatment non-adherence (a50); two predicted metastatic diagnoses while on treatment (a45, a65); one unique case showed fluctuating ctDNA results with no radiological evidence of disease despite a positive subsequent biopsy, resulting in patient-requested ctDNA cessation amidst recent positivity (a60).

Notably, two negative ctDNA cases showed unique results: one discovered contralateral recurrent disease despite two negative ctDNA tests (a11), and another showed new primary cholangiocarcinoma despite six negative ctDNA tests (a17).

Testing was also used to monitor metastases in two cases: 1 metastatic bone lesion with 8 negative draws that led to treatment cessation (a37); 1 oligometastatic case with 10 negative draws that led to treatment cessation (a42).

Discussion

Metastatic treatment monitoring without evidence of disease

Our study revealed several important uses of ctDNA in the clinical setting. An occasional, challenging clinical scenario includes patients who have initially presented with metastatic disease, completed aggressive treatments with curative intent, and are now without any evidence of disease. In our study, ctDNA was used to inform treatment response and help with decision-making in patients with negative ctDNA results following a metastatic breast cancer diagnosis (a37, a42: 2.3, Table 1). These persistently negative molecular results provided clinicians another tool to feel more confident that there was no evidence of disease and that treatment was optimally given. Together with traditional disease evaluation (CT CAP or PET/CT imaging and lab work, including tumor markers) without any evidence of disease, this allowed some rationale for treatment de-escalation for those participants (Figure 3). Both patients were under age 40 at the time of diagnosis, and quality of life issues were a significant concern with continued aggressive metastatic treatment regimens. Most studies assessing ctDNA’s role in treatment monitoring suggest that treatment response is best reflected in those who test ctDNA positive.¹⁹ While no studies have advocated for its efficacy in confirming treatment response in those with persistently negative ctDNA following treatment, in our study, ctDNA testing gave clinicians a chance to closely monitor patients longitudinally and further personalize treatment strategies to minimize the burden of aggressive treatment.

Table 1.

Unique breast cancer cases and trajectories.

ctDNA Result Category	Clinical Role of ctDNA	Case ID and Clinical Summary
Positive ctDNA findings
2.1 Positive ctDNA testing	Encouraging treatment adherence	(a50) A stage III, HR−/HER2+ left breast cancer s/p bilateral mastectomies, declined adjuvant chemotherapy; however, it agreed with HER2-directed therapy alone. After 7 months, she stopped HER2 treatment. A baseline ctDNA was negative; however, it became positive 3 months after treatment cessation. This positive ctDNA finding encouraged the patient to restart HER2 treatment, and a follow-up ctDNA test taken 3 months later was again negative.
	Predicting recurrent disease	(a65) A stage III, BRCA2+, HR+/HER2− right breast cancer s/p bilateral mastectomies, adjuvant chemotherapy and radiation, on anastrozole and CDK4/6 inhibitor (did not tolerate), had 2 negative ctDNA tests initially. After a positive ctDNA test prompted PET/CT imaging, which revealed a left 4th rib lesion-bx proven recurrence, s/p SBRT, systemic treatment changed to exemestane and zoledronic acid. Repeat ctDNA was negative 5 months later, and the patient remains without any evidence of disease.
	Predicting recurrent disease	(a45) A stage II TNBC completed NACT with chemotherapy and immunotherapy s/p lumpectomy with non-pCR, continued immunotherapy, capecitabine, and radiation. A baseline ctDNA test was positive and remained positive on subsequent testing. Metastatic disease was found in the liver 1 year later (12-month lead time).
	Predicting recurrent disease	(a05) A premenopausal stage II BRCA2+ HR+/HER2− s/p NACT and bilateral mastectomies, radiation, and tamoxifen (patient declined OFS + AI recommendation) had a negative baseline ctDNA test. However, subsequent testing turned positive 4 months later without radiological evidence of disease, encouraging a switch to OFS and AI, which was not tolerated. Five subsequent ctDNA tests were persistently positive, despite no radiographic evidence of disease until a sternal lesion was seen and biopsy confirmed metastatic recurrence (12-month lead time).
	Predicting recurrent disease	(a57) A perimenopausal, stage III HR+/HER2− breast cancer s/p left mastectomy, adjuvant chemotherapy, radiation, and tamoxifen + CDK4/6 inhibitor had a negative baseline ctDNA test, which then turned positive after 7 months, prompting a switch to an AI despite no radiological evidence of disease. Subsequent ctDNA levels continued to increase until radiographic evidence confirmed bone lesions (3-month lead time).
2.2 Fluctuating positive ctDNA	Predicting recurrent disease	(a52) A stage III TNBC treated with NACT-chemotherapy and immunotherapy, s/p L mastectomy with non-pCR RCB-2, radiation, and adjuvant capecitabine with immunotherapy had an initially negative ctDNA followed by fluctuating mildly positive tests while on adjuvant treatment, and a PET/CT without any evidence of disease. ctDNA was again negative when adjuvant treatment was completed, but became detectable 3 months later. An initial post-treatment PET/CT was negative, but the patient was then found to have peritoneal carcinomatosis on CT CAP 4 months after completing adjuvant treatment (11-month lead time).
	Predicting Recurrent Disease	(a60) A stage I, HR+/HER2− s/p lumpectomy, adjuvant chemotherapy, and radiation on anastrozole self-requested ctDNA testing from her PCP. Her initially negative ctDNA test became positive after 3 months with a negative PET/CT. However, the patient self-palpated a neck mass, and a biopsy confirmed local recurrence (3-month lead time). After local resection and radiation, treatment was changed to exemestane (declined CDK4/6 inhibitor), and a subsequent ctDNA test was initially negative but turned positive 3 months after the resection, during which a PET/CT showed no evidence of disease. Patient has declined all further ctDNA testing due to anxiety and continues to follow up with personalized alternative therapy.
Negative ctDNA findings
2.3 Negative ctDNA testing	Metastatic treatment monitoring without evidence of disease	(a37) A stage IV HR+/HER2− with bone lesions s/p palliative right mastectomy, chemotherapy, and radiation on AI, CDK4/6 inhibitor, and zoledronic acid. Bone lesions were initially stabilized, with some completely resolving. A PET/CT showed no evidence of disease for over 3 years, along with 8 negative ctDNA tests. The decision to stop the CDK4/6 inhibitor was made due to tolerability and overall no evidence of disease. She has no evidence of disease on AI treatment alone.
	Metastatic treatment monitoring without evidence of disease	(a42) A stage IV HR−/HER2+, oligometastatic left rib, s/p THP, SBRT to rib, was on adjuvant HP s/p lumpectomy-pCR and radiation. After a PET/CT showed no evidence of disease and 10 negative ctDNA tests, she remains on HP alone.
2.4 Negative ctDNA, however, with disease detected	New primary cholangiocarcinoma	(a17) A stage II HR−/HER2+ s/p NACT TCHP, lumpectomy, with non-pCR, who completed adjuvant TDM1 therapy, found an abnormal liver mass, difficult with biopsy, despite a negative baseline ctDNA test. Liver resection revealed a new stage I primary cholangiocarcinoma. Additional ctDNA tests remained negative. Currently unable to utilize ctDNA testing for cholangiocarcinoma given prior breast cancer history.
	Secondary recurrence	(a11) A right stage II TNBC s/p NACT without immunotherapy, s/p bilateral mastectomies with non-pCR, adjuvant chemotherapy, and radiation, baseline ctDNA test negative. However, 2 months after testing, clinical exam and imaging revealed a suspicious left axillary lesion, biopsy confirmed contralateral left TNBC without any distant evidence of disease. She underwent chemotherapy + immunotherapy, resection-non-pCR, radiation, and adjuvant capecitabine. Additional ctDNA tests remained negative, with imaging showing no evidence of disease.

AI, aromatase inhibitor; NACT, neoadjuvant chemotherapy; NH, Native Hawaiian; OFS, ovarian function suppression; pCR, pathological complete response; SBRT, stereotactic body radiation therapy; RCB-2, residual cancer burden index class 2 = moderate burden; THP, docetaxel + trastuzumab + pertuzumab chemotherapy; TNBC, triple-negative breast cancer.

Predicting recurrent disease and treatment upgrading

Our study demonstrated the role of ctDNA positivity in expediting metastatic diagnoses and exploring early treatment approaches, potential uses speculated by previous trials due to its negative prognostic value.^14,17 Diagnostic imaging was offered to those with positive ctDNA tests at any time point. This allowed patients and clinicians to consider potential treatment options sooner. Positive ctDNA testing helped guide earlier diagnostic imaging, which led to subsequent metastatic diagnoses in two cases (a45, a65: 2.1, Table 1). In one of these cases (a65), positive ctDNA testing prompted both a positive PET/CT and a treatment upgrade to exemestane and zoledronic acid, which led to ctDNA clearance within 5 months following (Figure 1). Early ctDNA clearance within 3 weeks has been shown to predict pCR, particularly in TNBC subtypes,¹⁷ suggesting that proactive interventions may delay or mitigate disease progression.

Figure 1.

Clinical trajectories of patients with positive ctDNA. (a50) Encouraging treatment adherence. (a65, a45, a05, and a57) Predicting recurrent disease change.

In two additional cases with ctDNA positivity (a05, a57: 2.1, Table 1), persistently positive ctDNA led to early treatment upgrades despite consequent negative imaging. This approach did not ultimately prevent metastatic disease (lead times of 12 and 3 months, respectively) (Figure 1), but since ctDNA has demonstrated lead times of at least 9 months,^15,16 our clinicians wanted to leverage ctDNA positivity to initiate optimized treatments earlier.

While current NCCN guidelines rely on clinical signs as the primary indication for imaging, ctDNA’s lead time may offer an additional way to detect asymptomatic metastases at a possibly more dormant phase. With the latest American Society of Clinical Oncology (ASCO) recommendations on the use of tumor markers dating back to 2015, long-term studies on detecting and treating recurrence earlier with ctDNA may discover new insights into how early detection and intervention affect patient outcomes.²³ Until this is established, ctDNA can still serve as a tool to characterize disease burden and provide clarity to patients with high-risk breast cancer, who often contend with the unpredictable nature of their disease progression.

Discrepant results

Our findings demonstrate both the potential and challenges associated with ctDNA in clinical practice. Although trials have demonstrated the high sensitivity of ctDNA,¹⁴ our study encountered several instances of fluctuating findings, such as discrepancies with radiological results and clinical symptoms. In two cases (a52, a60: 2.2, Table 1), ctDNA conflicted with radiological findings, turning positive despite subsequent negative PET/CTs, leading to considerable patient anxiety. In one case (a60), fluctuating ctDNA positivity amidst a negative PET/CT and pertinent clinical symptoms, which resulted in an eventual metastatic diagnosis, led the patient to cease subsequent ctDNA testing due to anxiety (Figure 2). For patients at high risk for recurrence, such conflicting results can be emotionally distressing, highlighting the importance of clinicians communicating the limitations of ctDNA and managing patients’ expectations accordingly.

Figure 2.

Clinical trajectories of fluctuating patients with positive ctDNA. (a52 and a60) Predicting recurrent disease.

Figure 3.

Clinical trajectories of patients with negative ctDNA. (a37 and a42) Metastatic treatment monitoring without evidence of disease.

Similarly, one case with negative ctDNA results (a11: 2.4, Table 1) initially presented with negative radiological and biopsy findings but water later found to have contralateral breast cancer recurrence (Figure 4). This emphasizes key areas for improving the clinical application of ctDNA in the real-world setting. Tumor-agnostic assays identify prevalent mutations independent of prior tumor mutations and typically require a sufficient tumor burden, often seen in metastatic disease, to yield detectable ctDNA levels. By contrast, tumor-informed assays are designed to track mutations specific to the original tumor. Although tumor-agonistic assays hold potential for detecting MRD in patients with breast cancer, their clinical utility remains under investigation.²⁴

Figure 4.

Clinical trajectories of patients with negative ctDNA with disease detected. (a17) New primary cholangiocarcinoma. (a11) Secondary recurrence.

Currently employed tumor-informed ctDNA tests remain inherently limited to monitoring mutations based on the original tumor and may not be able to detect recurrences arising in new locations or tissue, such as in our case of contralateral recurrence. As such, these assays would not be expected to detect a new primary malignancy, such as the cholangiocarcinoma in this case (a17: 2.4, Table 1). While this represents a fundamental limitation of tumor-informed assays, it also raises potential for misinterpretation among clinicians and patients, especially given that breast cancer is known to metastasize to external sites such as the bones, liver, and lungs.

To maximize the clinical utility of ctDNA testing, clinicians should ensure that patients understand the capabilities and limitations of these assays. Future research directions, potentially integrating tumor-informed assays with tumor-agnostic approaches, may enhance the sensitivity of recurrent disease detection, offering a more comprehensive strategy for early detection regardless of the original tumor’s biology.

Patient experience with ctDNA testing

While ctDNA poses a risk of discrepant results and associated anxiety, it can also promote treatment adherence, particularly in those undergoing treatment for earlier-stage cancers. In one case (a50: 2.1, Table 1), a patient undergoing ctDNA testing who failed to adhere to treatment responded to positive ctDNA tests by resuming treatment without the need of a diagnostic scan, after which a follow-up ctDNA turned negative (Figure 1). Patients with breast cancer endure substantial side effects of treatments, which reduce quality of life and make treatment adherence challenging. In our case, the patient did not opt for the optimal treatment regimen due to its daunting side effect profile, yet still struggled on trastuzumab alone. This case demonstrates how ctDNA testing can serve as a tangible reminder to patients of the importance of treatment persistence in high-risk scenarios where a pronounced risk of recurrence is not to be overlooked.

Overall, ctDNA can lead to earlier detection of recurrence and intervention, providing a tailored disease evaluation to patients and even promoting adherence in some cases. However, its consequences include the risk of unnecessary workup and treatment, especially in discrepant findings, which can especially burden patients emotionally. With a 30% recurrence rate,^2,3 patients with breast cancer who are in remission may experience uniquely heightened levels of anxiety, especially those at high risk. Studies have shown that breast cancer survivors living with cancer for at least 2 years exhibit significantly higher incidence of anxiety disorders than the general population, with 17.9% meeting clinical criteria for anxiety.²⁵ The emotional toll of scanxiety, characterized by distress and anxiety linked to cancer-related imaging and tests,²⁶ compounds these feelings. As ctDNA gains notoriety for its potential in detecting breast cancer recurrence, it may expand the repertoire of diagnostic tests available to patients with breast cancer. However, the possibility of discrepant results, as observed in our study, may worsen patients’ feelings of unease, mistrust, and panic. These very harms of serial ctDNA monitoring, in addition to increased diagnostic procedures and costs, unnecessary treatment upgrading, and heightened patient anxiety, support the current ASCO and National Clinical Trials Network guidelines, which do not endorse ctDNA testing for early-stage breast cancer.²¹ As radiological relapse is often used to validate ctDNA positivity, the experience of receiving liquid biopsies and imaging is closely intertwined, and conflict between the two can exacerbate anxiety as demonstrated in our study (a52, a60: 2.2, Table 1). Future studies should further characterize patients’ emotional experiences of ctDNA testing and develop interventions to address them.

Racial and ethnic disparities

Our study is inclusive of a wide range of ethnic and racial diversity without restriction to specific subtypes. The majority of our cohort consisted of Asian (45.45%), White (31.82%), and Native Hawaiian/Pacific Islander (NHPI; 16.67%) individuals, addressing significant gaps in breast cancer research (Table 2). NHPI individuals exhibit some of the highest cancer mortality rates,²⁷ and Hawaii breast cancer incidence rates remain the highest in the nation.²¹ National ctDNA databases have shown that patients who identify as Asian and NHPI utilizing ctDNA are the least represented (4.7%) compared to patients who identify as White (71.5%), Black (12.1%), and Hispanic (9.1%). In addition, the aggregation of these groups into broader AAPI categories²⁸ may obscure substantial health disparities among Pacific Islanders, with differences in 10-year survival outcomes as wide as 9% between NHPI and East Asian patients with breast cancer.²⁹ Studies investigating racial inequities of ctDNA similarly conflate NHPI populations with “Asian/other/unknown,” with similarly low participation (9.7%) compared to patients who identify as White (24.8%), Black (32.6%), and Hispanic (33%).³⁰ Aggregating Native Hawaiian and Asian American race categories could similarly impact reports on ctDNA’s negative prognostic utility, which has become evident in studies. Of the six cases in our study with positive ctDNA whose findings led to earlier radiological confirmation or intervention (2.1 and 2.2, Table 1), five were white. The one case that failed to detect contralateral recurrence was Filipino (2.4, Table 1). In larger studies, appropriate demographic reports are crucial to equitably advance this field of personalized medicine in diverse patient populations.

Table 2.

Demographic characteristics: Age, race/ethnicity, and breast cancer subtype.

Variable	Patients, No. (%)
Variable	Overall(N = 67)	HR+(N = 25)	HR+/HER2+(N = 7)	HR−/HER2+(N = 9)	TNBC(N = 26)
Age, years
20–29	1 (1.49%)	0 (0.00%)	0 (0.00%)	1 (11.11%)	0 (0.00%)
30–39	11 (16.42%)	4 (16.00%)	1 (14.29%)	3 (33.33%)	6 (23.08%)
40–49	6 (8.96%)	4 (16.00%)	3 (42.86%)	2 (22.22%)	6 (23.08%)
50–59	7 (10.44%)	8 (32.00%)	1 (14.29%)	2 (22.22%)	4 (15.38%)
60–69	8 (11.94%)	6 (24.00%)	2 (28.57%)	1 (11.11%)	6 (23.08%)
70–79	5 (7.46%)	2 (8.00%)	0 (0.0%)	0 (0.00%)	3 (11.54%)
80–89	1 (1.49%)	1 (4.00%)	0 (0.00%)	0 (0.00%)	1 (3.85%)
Ethnicity
Black	2 (3.03%)	0 (0.00%)	0 (0.00%)	0 (0.00%)	2 (7.69%)
Chinese	5 (7.58%)	0 (0.00%)	2 (28.57%)	1 (11.11%)	2 (7.69%)
Filipino	7 (10.61%)	3 (12.0%)	0 (0.00%)	0 (0.00%)	4 (15.38%)
Hispanic	3 (4.55%)	2 (8.00%)	0 (0.00%)	1 (11.11%)	0 (0.00%)
Japanese	15 (22.73%)	5 (20.0%)	2 (28.57%)	1 (11.11%)	7 (26.92%)
Korean	2 (3.03%)	0 (0.00%)	0 (0.00%)	0 (0.00%)	2 (7.69%)
Native Hawaiian or Pacific Islander	11 (16.67%)	5 (20.0%)	2 (28.57%)	2 (22.22%)	2 (7.69%)
Vietnamese	1 (1.52%)	0 (0.00%)	0 (0.00%)	0 (0.00%)	1 (3.85%)
White	21 (31.82%)	10 (40.0%)	1 (14.29%)	4 (44.44%)	6 (23.08%)

TNBC, triple negative breast cancer.

Study limitations

While this study provides valuable real-world insights, it is not without limitations. The early-access program allowed clinicians to be more experimental than systematic with ctDNA orders, which led to earlier detection and intervention in several cases. No ctDNA timepoints were required beyond baseline assessments, and collections were not randomized. In addition, baseline ctDNA tests were established throughout different stages of the disease course. Nevertheless, studies involving patients with high-risk breast cancer have previously established ctDNA’s negative prognostic capability at any time point, whether persistent positivity occurs before, during, or after neoadjuvant therapy.¹⁷ Our relatively small sample size and lack of randomization constrain the generalizability of our findings across broader populations and limit our ability to confirm our positive ctDNA results and earlier intervention as a surrogate for improvement in overall survival, a gap yet to be addressed in ongoing trials. However, our results showcase how ctDNA has been used in real-life clinical situations, affecting both patient and clinician decision-making, ultimately showcasing its utility and harms.

Future directions

While persistent ctDNA positivity has consistently predicted recurrence,^17,19 it is important to acknowledge ongoing criticism that the earlier detection of molecular residual disease has not been shown to improve overall survival in metastatic breast cancer. This concern reflects a broader skepticism toward intensive surveillance strategies, particularly when interventions are unlikely to alter the overall disease trajectory and thus may introduce potential harms, including limited therapeutic options later in the disease course, increased healthcare costs, and heightened patient anxiety.²⁰

Ongoing trials, such as the SURVIVE study (NCT05658172), are currently assessing potential survival benefit in a surveillance arm utilizing ctDNA in patients with intermediate-to-high-risk breast cancer receiving follow-up care.³¹ Although survivorship bias has not yet been clinically proven in ctDNA studies, machine learning models have been validated in external cohorts using ctDNA metrics, indicating the potential of ctDNA to impact risk stratifications and thus treatment decisions. Initial predictive models for survival integrate factors such as the number of detected variants and the total cell-free DNA extracted from plasma, posing future avenues to further explore in larger cohorts and trials.³²

While our study does not address survival benefit directly, it highlights that ctDNA can still meaningfully guide clinical decision-making, particularly in cases where persistent ctDNA negativity may support treatment de-escalation and improved quality of life in patients facing the psychological and physical toll of metastatic treatment. These real-life clinical cases highlight that utility may not only lie in prolonging survival, but also in enabling patient-centered, individualized care in the metastatic setting.

Summary

Our findings represent a real-world evaluation of clinician practices in response to testing as well as patient experience, demonstrating ctDNA’s potential and its challenges. Nevertheless, our study showed that ctDNA can support clinicians in personalizing treatment regimens and improving treatment adherence in patients with high-risk breast cancer. This testing can serve as a valuable tool in managing high-risk breast cancer with appropriate discussion of limitations and clear communication of both its potential benefits and harms.

Conclusion

Our findings demonstrate cases where ctDNA serves as a valuable tool for disease evaluation. In some cases, earlier recurrence detection in patients with persistently positive ctDNA allowed clinicians to perform proactive treatment upgrades. In addition, ctDNA negativity was used to monitor metastatic treatment response and the potential for de-escalation. While ctDNA can characterize disease burden and facilitate earlier diagnosis of recurrence and potential intervention, it is unclear whether this ultimately changes the disease course outcome. There is a need for more research to refine its clinical applications. While the emotional burdens of positive ctDNA results have been seen in this study, ctDNA positivity has also demonstrated potential in aiding treatment adherence. Future studies are ongoing to discern whether early treatment of molecular breast cancer relapse delays or prevents subsequent clinical relapse. This study also emphasizes the need to include diverse populations in future ctDNA studies and the importance of considering the emotional impact of ctDNA testing on patients.

Footnotes

Acknowledgements

We would like to thank the patients who participated in this study.

Declarations

ORCID iDs

Kasen Wong

Jami Aya Fukui

References

American Cancer Society. Cancer facts and figures 2022, https://www.cancer.Org/research/cancer-facts-statistics/all-cancer-facts-figures/cancer-facts-figures-2022.html (2022, accessed 24 January 2025).

Cheng

Swartz

Zhao

, et al. Hazard of recurrence among women after primary breast cancer treatment—a 10-year follow-up using data from SEER-Medicare. Cancer Epidemiol Biomarkers Prev 2012; 21(5): 800–809.

Noone

Howlader

Krapcho

, et al. SEER cancer statistics review, 1975–2015. National Cancer Institute, https://seer.cancer.gov/archive/csr/1975_2015/index.html (2018, accessed 24 January 2025).

National Cancer Institute. Recurrent cancer: when cancer comes back, https://www.cancer.gov/types/recurrent-cancer#:~:text=Regional%20recurrence%20means%20that%20the,called%20metastasis%20or%20metastatic%20cancer (2020, accessed 24 January 2025).

DeSantis

Siegel

Bandi

, et al. Breast cancer statistics, 2011. CA Cancer J Clin 2011; 61(6): 409–418.

National Cancer Institute. Cancer stat facts: female breast cancer, https://seer.cancer.gov/statfacts/html/breast.html (2024, accessed 24 January 2025).

Fillon

Breast cancer recurrence risk can remain for 10 to 32 years. CA Cancer J Clin 2022; 72(3): 197–199.

National Comprehensive Cancer Network. Breast Cancer, https://www.nccn.org/login?ReturnURL=https://www.nccn.org/professionals/physician_gls/pdf/breast.pdf (2023, accessed 24 January 2025).

De Mattos-Arruda

Siravegna

. How to use liquid biopsies to treat patients with cancer. ESMO Open 2021; 6(2): 100060.

10.

Miller

Schwartzberg

Precision medicine for breast cancer utilizing circulating tumor DNA: it is in the blood. Curr Treat Options Oncol 2020; 21(11): 89.

11.

Corcoran

Chabner

BA.

Application of cell-free DNA analysis to cancer treatment. N Engl J Med 2018; 379(18): 1754–1765.

12.

Lopez

Rassa

, et al. Direct comparison of circulating tumor DNA sequencing assays with targeted large gene panels. PLoS One 2022; 17(4): e0266889.

13.

Davidson

Croessmann

Park

BH.

The breast is yet to come: current and future utility of circulating tumour DNA in breast cancer. Br J Cancer 2021; 125(6): 780–788.

14.

Shaw

Page

Ambasger

, et al. Serial postoperative ctDNA monitoring for early detection of breast cancer recurrence. J Clin Oncol 2022; 40: 16s.

15.

Coombes

Page

Salari

, et al. Personalized detection of circulating tumor DNA antedates breast cancer metastatic recurrence. Clin Cancer Res 2019; 25(14): 4255–4263.

16.

Yang

Zhang

J-Y

, et al. The ctDNA-based postoperative molecular residual disease status in different subtypes of early-stage breast cancer. Gland Surg 2022; 11(12): 1924–1935.

17.

Magbanua

MJM

Brown

Swigart L

Ahmed

, et al.; c-TRAK TN Investigators. Clinical significance and biology of circulating tumor DNA in high-risk early-stage HER2-negative breast cancer receiving neoadjuvant chemotherapy. Cancer Cell 2023; 41(6): 1091–1102.e4.

18.

Turner

Swift

Jenkins

, et al. Results of the c-TRAK TN trial: a clinical trial utilising ctDNA mutation tracking to detect molecular residual disease and trigger intervention in patients with moderate- and high-risk early-stage triple-negative breast cancer. Ann Oncol 2023; 34(2): 200–211.

19.

Loi

Johnston

Arteaga

, et al. Prognostic utility of ctDNA detection in the monarchE trial of adjuvant abemaciclib plus endocrine therapy (ET) in HR+, HER2−, node-positive, high-risk early breast cancer (EBC). J Clin Oncol 2024; 42(17_suppl): LBA507.

20.

Puszati

Rozenblit

Balancing the benefits and harms of ctDNA surveillance in early-stage breast cancer. Clinical News from the American Society of Clinical Oncology, https://dailynews.ascopubs.org/do/balancing-benefits-and-harms-ctdna-surveillance-early-stage-breast-cancer (2024, accessed 24 January 2025).

21.

National Cancer Institute. State cancer profiles, https://www.statecancerprofiles.cancer.gov/ (2024, accessed 24 January 2025).

22.

Gagnier

Kienle

Altman

, et al. The CARE guidelines: consensus-based clinical case reporting guideline development, https://www.equator-network.org/reporting-guidelines/care/ (2013, accessed 14 April 2025).

23.

Riggio

Varley

Welm

AL.

The lingering mysteries of metastatic recurrence in breast cancer. Br J Cancer 2021; 124: 13–26.

24.

Panet

Papakonstantinou

Borrell

, et al. Use of ctDNA in early breast cancer: analytical validity and clinical potential. NPJ Breast Cancer 2024; 10(1): 50.

25.

Mitchell

Ferguson

Gill

, et al. Depression and anxiety in long-term cancer survivors compared with spouses and healthy controls: a systematic review and meta-analysis. Lancet Oncol 2013; 14(8): 721–732.

26.

Derry-Vick

Heathcote

Glesby

, et al. Scanxiety among adults with cancer: a scoping review to guide research and interventions. Cancers 2023; 15(5): 1381.

27.

O’Hare

WP.

Census coverage of the Native Hawaiian or Pacific Islander population. In: O’Hare

, ed. Differential undercounts in the U.S. census. SpringerBriefs in population studies. Cham: Springer, 2019, pp. 109–115.

28.

Davis

Bucheit

Zhang

, et al. Utilization of circulating tumor DNA (ctDNA) testing by race and ethnicity in more than 135,000 patients. J Clin Oncol 2024; 42: 16s.

29.

Taparra

Fukui

Killeen

, et al. Racial and ethnic disparities in rates of invasive second breast cancer among women with ductal carcinoma in situ in Hawai’i. JAMA Netw Open 2021; 4(10): e2128977.

30.

D’Aiello

Narvel

Daylan

, et al. Investigating racial inequities of circulating tumor DNA (ctDNA) use in patients with non-small cell lung cancer: a real world analysis. J Clin Oncol 2023; 41: 16s.

31.

Mergel

Huesmann

Friedl

, et al. SURVIVE study—a multicenter, randomized, controlled phase 3 superiority trial, evaluating liquid biopsy guided intensified follow-up surveillance in women with intermediate-to high-risk early breast cancer. Paper presented at Cancer Research, San Antonio Breast Cancer Symposium, 5–9 December 2023, vol. 84, no. 9, Paper No. PO1-20-05, TX, USA.

32.

Assaf

ZJF

Zou

Fine

, et al. A longitudinal circulating tumor DNA-based model associated with survival in metastatic non-small-cell lung cancer. Nat Med 2023; 29: 859–868.

Exploring the utility of ctDNA testing in high-risk breast cancer patients in a community setting: case series