HPV-Positive Oropharyngeal Cancers: Predicting Treatment Response Using ddPCR

During this year’s AACR Annual Meeting we spoke to Julianna Supplee, Research Technician at the Robert and Renée Belfer Center for Applied Cancer Science at Dana-Farber Cancer Institute, to learn more about the team’s most recent study on the use of plasma HPV cell-free DNA as an early predictor of treatment response in patients with advanced oropharyngeal cancer. Julianna discusses the use of droplet digital PCR (ddPCR) as a means for confirming treatment response. She highlights the potential advantages of ddPCR over imaging techniques, as well as efforts to optimize ddPCR for use in the clinic.

Laura Mason (LM): Could you provide a brief summary of the study you presented during the Annual AACR 2018 Meeting?

Julianna Supplee (JS): Our study is on HPV-positive oropharyngeal cancers. We developed a droplet digital PCR (ddPCR) assay against viral genes from HPV to see if we can detect and quantify viral DNA in plasma, and then compared the HPV cell-free (cf)DNA concentration, throughout treatment at various timepoints, with the clinical parameters of those patients.

LM: Could you tell us a little more about oropharyngeal squamous cell carcinomas (OPSCC)?

JS: OPSCC is not particularly common, although the incidence of these cancers is rising in the US. It's a type of cancer that originates in the squamous cells, usually in the neck or in the back of the throat. About three quarters are directly caused by the human papillomavirus (HPV). For the most part it's caught early enough that you can treat it aggressively, and most patients do respond well. However, there is a subset of HPV⁺ OPSCC patients (approximately 10–15%) that experience disease recurrence and go on to develop advanced disease. Unfortunately, there are no specific targeted therapies for these patients. You can cycle through standard chemo, radiation, and surgery -- but these recurring patients unfortunately usually succumb.

LM: What were the key findings of the study?

JS: Importantly, we were able to detect HPV DNA in plasma with ddPCR. To our knowledge, that's never been done before. There's been some serum studies, but not with ddPCR, so that was exciting. We didn't really know what to expect when we got the samples.

First, we looked at the baseline plasma levels and found that patients with very low or undetectable HPV DNA levels had local disease. In comparison, we saw plasma levels really shoot up in patients with distant metastasis. If they had both (local and distant disease), it was even higher.

We got six draws from each patient over the course of about three months. The patient would also get regular scans. When plasma levels dropped, it corresponded with a good scan, and when the plasma levels rose, it corresponded with a bad scan.

What was truly impressive was that the plasma changed before the imaging, so in a responder you would see the plasma levels change approximately two weeks before the imaging. And if they progress, you could see progression in plasma almost a whole month ahead of time.

LM: What are the benefits of using ddPCR to confirm treatment response, compared to imaging techniques?

JS: Just practically speaking, getting a tube of blood drawn is so much easier than going in for a CT scan or a PET scan, and a tumor biopsy for that matter. When it comes to location, it is more practical too, because patients can go to their local clinic and get their blood drawn. The patient wouldn’t need to find a specialized center that could potentially be extremely far away.

We are working on optimizing ddPCR to make it more translatable to the clinic. Hopefully once we get ddPCR into the clinic, it will be a better, faster, and more cost-effective approach.

LM: Your abstract mentions ‘further studies’ could you expand on this? Is this research being carried out by yourself/ your team?

JS: We are working on integrating ddPCR into already-existing or upcoming clinical trials. Again, the focus is for oral cancer associated with HPV.

In the current study, our cohort was smaller and a more diverse assortment of people. We didn't have any restrictions on what kind of treatments the patients were going through.

They weren't treatment-naive, but rather we were just following them and seeing if their plasma results made sense compared to the clinical results, and they did. Our results were really impressive considering it was a heterogeneous patient pool. If you went on a case-by-case basis and looked at each clinical picture, their plasma results made perfect sense when you considered all the clinical parameters. So hopefully with a bigger sample pool and a more homogeneous treatment regiment, we'll be able to tease out how much of the variables might affect the trends we're seeing.

Julianna Supplee was speaking to Laura Elizabeth Mason, Science Writer for Technology Networks.