Liquid Biopsy May Answer One of the Most Important Questions in Cancer Treatment

Liquid biopsy is a biotechnology that allows DNA from cancer cells to be detected in bodily fluids; blood is the bodily fluid most commonly used for this purpose.

Liquid biopsy is playing an increasingly critical role in oncology treatment decisions. At present, this method has sufficient evidence and clinical validity to map the tumor's genetic makeup in order to determine eligibility for targeted drugs directed at molecular targets, in already diagnosed advanced cancers, and to identify cancer-related mutations.

However, the use of circulating tumor DNA (ctDNA) measurement for early diagnosis or identifying patients at high risk of cancer recurrence after surgery has not received FDA (US Food and Drug Administration) approval to date.

Liquid biopsy has great potential for predicting residual disease and the risk of recurrence (relapse) after primary treatments such as surgery and radiotherapy in early and locally advanced solid (organ) cancers.

Several studies across multiple tumor types presented at the ASCO 2022 Congress show that detecting circulating tumor DNA after primary treatment strongly predicts subsequent disease recurrence.

These studies may help oncologists use circulating tumor DNA to make more informed decisions about risk-adapted adjuvant (protective treatment after primary treatment) approaches, using it as a biomarker.

Liquid biopsy has enormous potential to monitor the presence and burden of cancer by detecting circulating tumor DNA (ctDNA) through a simple blood draw. Interest in using ctDNA as a cancer biomarker is gaining momentum in clinical and research settings, and developments in this field prompted the FDA (US Food and Drug Administration) to publish guidance on how to use ctDNA testing when developing drugs for early-stage tumors.

The guidance describes 3 potential uses of ctDNA in treatment decision-making:

1. Selecting patients eligible for targeted cancer drugs based on molecular alterations;
2. Monitoring tumor response at the molecular level through minimal residual disease to determine recurrence risk; and
3. Predicting long-term disease outcomes early and arranging patient follow-up accordingly

Several studies presented during the 2022 ASCO Annual Meeting in multiple tumor types, including head and neck squamous cell cancer (Abstract 6017), non-small cell lung cancer (Abstract 8534), breast cancer (Abstract 103), soft tissue sarcoma (Abstract 11547), and oral cancer (Abstract 6006), focused on using liquid biopsy to identify residual disease after primary treatment.

Being ctDNA-negative does not mean you are cured, but being ctDNA-positive means you are not cured.

Taken together, the findings demonstrate the power of ctDNA detection to herald disease progression after curative treatment—in many cases, months before recurrence becomes radiologically or clinically apparent...

Studies Supporting Risk-Adapted Treatment

Two studies in particular demonstrated the prognostic potential of ctDNA (its ability to predict the course of disease).

In the first study, Dr. Susanne Flach of the University of Munich and colleagues conducted a single-center prospective cohort study (LIONESS) in patients with stage I-IVB head and neck squamous cell cancer.

They conducted this study to determine whether postoperative ctDNA detection could serve as a biomarker for residual tumor and to assess the potential of ctDNA to detect early molecular recurrence before clinically confirmed recurrence (Abstract 6017).

Data from 46 patients followed with serial blood ctDNA monitoring revealed markedly different outcomes based on postoperative ctDNA status. Most notably, 69% (11/16) of individuals who tested positive for ctDNA after surgery experienced clinical recurrence, whereas cancer recurrence was observed in 0% of those found to be ctDNA-negative (0 out of 30) ( P = 0.0000002).

Among the 11 people who experienced clinical recurrence, ctDNA was detected a median of 122 days (range: 1-260) before clinical progression.

The second study, led by Soyeong Jun of Stanford University and Nikhil Shukla of Indiana University, was an analysis of the phase-II BTCRC LUN 16-081 trial involving patients with stage 3 non-small cell lung cancer who received consolidation nivolumab with or without ipilimumab after chemoradiotherapy (Abstract 8534). Monitoring ctDNA throughout consolidation treatment showed that patients with detectable ctDNA (that is, ctDNA-positive patients) before, during, or after immunotherapy had shorter progression-free survival (PFS). Conversely, 2-year PFS rates were significantly higher in patients with declining or undetectable ctDNA levels after 1 cycle of consolidation than in patients with rising ctDNA levels (24-month PFS: 0% versus 73%, P < 0.0001), and 91% of patients with undetectable ctDNA at the end of consolidation were still alive 2 years after treatment.

These studies show that we may be starting to accurately select patients who are likely to need more treatment. If patients are still ctDNA-positive after their planned treatments, perhaps we should then plan salvage treatments.

The team is currently conducting a prospective clinical trial that intensifies treatment by adding chemotherapy to immunotherapy in patients without a ctDNA response after several cycles of immunotherapy. ctDNA analysis may enable personalized treatment by facilitating treatment escalation in nonresponders and/or helping determine the optimal duration of immunotherapy in responders.

Using ctDNA to Predict the Long-Term Course of Disease

Post-treatment ctDNA positivity can vary considerably across tumor types, as demonstrated in the CHiRP study led by Dr. Marla Lipsyc-Sharf of the Dana-Farber Cancer Institute. CHiRP researchers used ctDNA monitoring to determine whether detection of minimal residual disease was associated with late recurrence events in patients diagnosed at least 5 years earlier with high-risk stage II/III hormone receptor-positive, HER2-negative breast cancer (Abstract 103).

Of the 83 patients, 8 (10%) had measurable ctDNA in blood samples taken 4.9 to 20 years after diagnosis. Notably, 6 of the 8 people with detectable ctDNA developed metastatic recurrence, with a median lead time of 12.4 months before clinical recurrence.

We may need to follow people for longer and adapt screening protocols to the biology of the cancer.

Of note, routine computed tomography (CT) surveillance is not performed in this patient group because the likelihood of recurrence is very low. It remains to be determined whether ctDNA monitoring will offer a feasible, cost-effective alternative for identifying molecular recurrence and allowing early treatment intervention.

Dr. Lipsyc-Sharf said, “Under current guidelines, patients in our study were followed clinically for recurrence, but were not assessed with concomitant body imaging in addition to plasma ctDNA testing.” “In future studies, it will be important to understand how many ctDNA-positive patients without known clinical recurrence have scan-detectable metastatic disease despite having no evidence of metastatic disease on imaging.”

Dr. Lipsyc-Sharf also noted that “prospective clinical trials are ongoing and that they plan to investigate whether intervention after ctDNA detection improves outcomes in patients without scan-detectable metastatic disease.”

The Power of ctDNA in Rare Tumor Types

With funding support from an ASCO Young Investigator Award, Abdulazeez Salawu of the Princess Margaret Cancer Centre in Toronto and colleagues conducted a pilot study on the feasibility of ctDNA testing in 10 patients with early-stage soft tissue sarcoma (STS) at high risk of recurrence (Abstract 11547).

The study met its primary endpoint by demonstrating the ability to detect ctDNA in 70% of patients with high-risk STS at diagnosis. After surgery, all patients were negative for ctDNA. ctDNA was detected during follow-up in 2 patients who subsequently experienced metastatic recurrence.

Dr. Salawu said, “Given that STS is a clinically and molecularly heterogeneous group of cancers, the personalized, tumor-informed, histology-independent approach to ctDNA detection used in our study is highly appropriate,” “Our results are promising for developing a much-needed biomarker to guide treatment and improve overall outcomes in this heterogeneous patient group, and this work is ongoing.”

This ctDNA concept is broadly relevant and may be useful in rare cancers for accelerating drug development. ctDNA could be incorporated into routine practice both to try to cure more patients by adapting treatment to risk and to introduce study designs in which ctDNA negativity may be a useful post-treatment follow-up marker instead of potentially longer follow-up. Clinical trials could thus be conducted more easily in rare cancers with fewer patients.

Tracking the Circulating Tumor: A Bespoke Test for the Patient's Own ctDNA

Most of the ctDNA studies described relied on personalized ctDNA assays developed using whole-exome sequencing data obtained from each person's tumor tissue—in other words, detection of DNA from the person's own cancer cells.

This is a highly precise and powerful detection method. However, it requires the tumor to undergo another genetic test after pathological examination, and the tumor must be sufficiently intact to actually generate the personalized sequencing data that can then be used to monitor the blood. This is a potential limitation in some tumor types.

In an analysis of the NRG-HN002 trial in patients with p16-positive oral cancer, an HPV biomarker, Dr. Sue S. Yom of the University of California, San Francisco, and colleagues took a different approach to ctDNA monitoring by focusing on the disease-causing HPV (Abstract 6006). More specifically, the research team monitored tumor tissue-modified viral HPV DNA circulating in the blood before, during, and after radiotherapy or chemoradiotherapy.

Dr. Yom said, “Tumor tissue-modified viral HPV does not require testing large amounts of tumor tissue.” “It is also less expensive and more efficient than ctDNA assays. The drawback, of course, is that it is specific to detecting high-risk HPV, which is not present in all head and neck cancers.”

Tracking 1 tumor-derived marker in a particular patient group is much more convenient and cost-effective than creating personalized ctDNA panels, and the approach appears to work. Of course, a highly tumor-specific biomarker must be found.

NRG-HN002 researchers reported 3 important findings. First, disease burden (that is, T and N stage) was weakly to moderately correlated with high pretreatment levels of tumor tissue-modified viral fragments. Second, chemoradiotherapy led to a more rapid decline in tumor tissue-modified viral fragments 2 weeks after treatment compared with radiation alone. Third, undetectability of tumor tissue-modified viral fragments correlated with 2-year outcomes. Specifically, 93.3% to 95.0% of patients with undetectable viral fragments after treatment experienced no regional cancer recurrence or disease progression throughout 2 years of follow-up.

The Future of ctDNA Testing

Given the favorable outcomes associated with undetectable ctDNA and the less favorable outcomes associated with detectable ctDNA after definitive treatment, the next step in improving cancer treatment success is to act on this information and intervene with additional treatment when appropriate. This step may involve giving adjuvant treatment to patients at high risk of progression after definitive treatment, or intensifying treatment for patients who remain at high risk of progression despite adjuvant treatment.

I think the main message across all these abstracts is the growing suggestion that ctDNA is increasingly likely to become a new biomarker we use to think about how to adapt adjuvant guidelines.

*

The uses and approval status of liquid biopsies in cancer can be seen below: