As precision medicine accelerates, breakthroughs in AI-driven genomics and GPU-accelerated computing are transforming how clinicians detect residual disease and plan post-surgical cancer treatment. What once required weeks of genomic analysis can now be completed in days, enabling physicians to incorporate molecular evidence into critical care decisions far earlier in the recovery cycle.
In this edition of the HealthTech Insights Top Voice Interview Series, Sudipto Ghosh, Global Head of Marketing and Sales at Intent Amplify, speaks with Wendy Winkler, Chief Scientific Officer at Droplet Biosciences, a company pioneering lymphatic fluid–based diagnostics for rapid detection of minimal residual disease (MRD).
The conversation explores how NVIDIA’s Parabricks platform is dramatically accelerating genomic sequencing workflows, allowing MRD signals to be detected during the pre-adjuvant therapy window. Wendy explains how earlier molecular insight could reshape oncology decision-making, improve treatment personalization, and help clinicians balance therapy intensity with patient outcomes.
As NVIDIA’s GPU-accelerated genomics platforms enable lymphatic fluid–based cancer detection within 24 hours of tumor removal, how should healthcare providers reassess post-surgical treatment planning timelines?
Wendy Winkler : To clarify, the patient sample is collected within 24 hours of tumor removal. Today, that sample is sent for laboratory testing, and lab results are returned to the physician in about 10 days. Approximately half of that time is sample processing including next-generation sequencing. Once the sequencing is complete, the other half of the time is the data processing and analysis of the sequencing results, which takes an additional 5 days. NVIDIA’s GPU platform dramatically improves the time to complete the data analysis, reducing that from 5 days to 1-2 days; thereby reducing total testing time to with one week of sample collection.
Today, under NCCN guidelines for HPV(-) head and neck squamous cell carcinoma, clinical teams use surgical pathology results to select a patient’s adjuvant treatment regimen. Adjuvant treatment (which may include radiotherapy, chemotherapy, and / or immunotherapy) begins by four to six weeks after surgery, generally before blood-based minimal residual disease (MRD) tests results are available. Droplet Biosciences has demonstrated that if you shift the biological sample to post-surgical lymphatic fluid and pair it with accelerated computing, you can return MRD results on a similar timeframe to pathology, early enough for them to be incorporated into that adjuvant therapy decision.
With NVIDIA Parabricks reducing sample processing and analysis time from 10 days to under seven – MRD results can be reported during the pre-adjuvant surgical recovery window. It allows providers to incorporate molecular evidence of residual disease into adjuvant planning earlier, ultimately leading to better patient outcomes.
With NVIDIA Parabricks reducing genomic alignment time from over 36 hours to under three, what operational changes must clinical labs implement to safely integrate AI-accelerated sequencing into surgical recovery workflows?
Wendy Winkler : LymphDetect test is offered as a Laboratory Developed Test in Droplet’s CLIA-certified lab, and as such, follows rigorous protocols for the validation of all lab techniques and algorithms employed in the test prior to any clinical use. Additionally, control samples are run through the lab and analysis pipeline with each sequencing batch to ensure expected performance, and successful proficiency testing of the complete specimen receipt -> report process is documented biannually.
When the NVIDIA Parabricks alignment tool was first implemented at Droplet, the team demonstrated that the alignment results were consistent with output from the slower, established pipeline in side-by-side comparisons across multiple samples.[WW3] Development and validation were then performed to make sure the Parabricks tool was compatible with the computer environment to ensure security and accuracy.
As AI-driven sequencing now allows detection of minimal residual disease during the same hospital admission, how might earlier genomic insight reshape oncologists’ decisions around adjuvant therapy?
Wendy Winkler : Earlier insight expands optionality. If residual disease can be detected within a week after surgery, at about the same time as traditional pathology results on the tumor arrive, oncologists gain actionable molecular information during the adjuvant decision window.
For patients, this means adjuvant therapy can be tailored using both pathology (studying tumor and lymph nodes that were removed) and well as molecular residual disease (evaluating what was left behind). This can enable intensified treatment for those who need it most. For those without detectable disease, clinicians may have stronger evidence to avoid overtreatment. Delivering results shortly after surgery provides peace of mind sooner and helps influence better outcomes overall.
To participate in our interviews, please write to our HealthTech Media Room at news@intentamplify.com
