Checking Your Answers: Monitoring Immunotherapy Effectiveness with ddPCR

When oncologists decide to use immunotherapy to treat a patient, they must make crucial determinations: Do the benefits outweigh the risks? And is it the best available option for the patient?

Although immunotherapies are effective in treating multiple types of cancers, including those that are often resistant to chemotherapy and radiation, they may not be effective in a significant fraction of patients who possess particular tumor types. For instance, checkpoint inhibitors, the most commonly used immunotherapy, can fail in up to 85 percent of patients treated for certain cancer types. Immunotherapies fail so often mainly because of our imperfect ability to accurately predict who will benefit.

When immunotherapy drugs do work, they can result in deep, long-lasting positive responses. But because of their unique ability to stimulate the immune system, they can induce a number of severe immune-related adverse events (irAEs) resulting from systemic inflammation. This puts further responsibility on physicians to ensure that a patient is on the right immunotherapy. 

Immunotherapy’s limited success rates, paired with its tendency to induce severe irAEs and the inaccuracy of predictive tests, may cause many patients more harm than good. Consequently, once treatment is started, oncologists should monitor their patients’ responsiveness to therapy and determine as early as possible whether or not the treatment is working. 

The current standard in assessing a patient’s response to therapy is the use of computed tomography (CT) imaging or magnetic resonance imaging (MRI). However, imaging-based approaches may not tell the whole story. For instance, up to 17 percent of patients with urothelial cancer, 15 percent of patients with renal cancer, and 8.3 percent of patients with melanoma who receive immunotherapy experience pseudo-progression. This is where a scan shows that a tumor has grown, when in reality it’s just temporarily inflamed from an influx of lymphocytes as a result of an effective immunotherapy treatment that will eventually cause it to shrink. Oncologists might not be able to distinguish between pseudo-progression and true progression for more than three months, making it challenging to decide whether the treatment was successful. Incorrectly mistaking pseudo-progression for true progression might put patients at risk of ceasing the use of an effective drug.

A promising alternative to imaging is liquid biopsy, which detects circulating cell-free tumor DNA (ctDNA) shed from dying tumor cells into the blood. Liquid biopsy can detect changes in ctDNA that reflect treatment effectiveness, or lack thereof, within weeks of starting treatment. It can also distinguish true progression from pseudo-progression. This approach is under active investigation using Droplet Digital PCR (ddPCR).

Unlike other phenotypic markers used as surrogates for tumor response to therapy, such as proteins or RNAs, ctDNA in blood or other bodily fluids is both a genetic and a phenotypic marker of therapy effectiveness. It measures the direct effects of the therapy on tumor cells, such as tumor cell turnover. ctDNA is thus a very specific biomarker for the tumor cells and can efficiently tell an oncologist if he or she has made the right therapeutic choice for the patient.

In one recent case, Australian researchers developed a ddPCR-based liquid biopsy to track BRAF and NRAS mutations in patients with stage IV melanoma to distinguish between responders and non-responders to anti-PD-L1 (+/- anti-CTLA-4) therapy. They found that longitudinal monitoring of ctDNA was an effective means to identify patients who did or did not respond well to the checkpoint inhibitors. Patients with persistently elevated ctDNA while on the therapy had poor prognosis, regardless of whether ctDNA was detectable at baseline or not. 

They followed this with a study of ctDNA monitoring to identify pseudo-progression in patients identified by CT scan as progressors. They found that ctDNA monitoring enabled pseudo-progression to be identified as an artifact of imaging scan misinterpretation, allowing for early correction of progression misdiagnosis.  

In another recent study on non-small cell lung cancer patients, researchers in the Netherlands used ddPCR-based liquid biopsy to measure KRAS exon 2 hotspot mutation levels in patients with NSCLC undergoing nivolumab treatment. They detected a positive response to anti-PD-L1 treatment, as characterized by distinctive kinetic ctDNA profiles, beginning just one week after the start of therapy, with further support for this response from measurements made within the first three to seven weeks. 

Liquid biopsies do harbor some limitations, but these are minor compared to the limitations of current pre-treatment predictive tests. For instance, liquid biopsies may not detect some tumors in locations where they do not slough ctDNA into the blood (e.g. brain metastases). Nonetheless, this plasma monitoring paradigm of immunotherapy effectiveness is a significant step forward as it removes many of the assumptions that must be made when predicting immunotherapy effectiveness with currently available pre-treatment prediction methods. ddPCR enables a short turnaround time and can be used to monitor treatment response more frequently and earlier than CT scans, as it is less invasive than a traditional biopsy and more cost-effective than imaging.

Because of their sensitivity, specificity, and speed, ddPCR-based liquid biopsies could provide an early and reliable measure of treatment effectiveness. This would help oncologists make smarter treatment decisions faster, sparing their patients from unnecessary side effects and saving precious time from being wasted on ineffective treatment regimens.