You’re looking at a disease that’s already being treated rather well but raises certain questions. Why were you so interested in looking at cell-free DNA?

Although Hodgkin lymphoma is a disease which is well treatable and highly curable today, there are a number of important issues that remain. First of all, for those patients who will be cured successfully by current therapy regimens, there is the important concern of late treatment related toxicities which compromise the long-term prospects of these cases. So there is an on-going need, there is still a pressing need, to de-escalate therapy in those patients who could benefit from that. But it’s very difficult to identify those cases.

Then there is another point: although the therapy of Hodgkin lymphoma has been a success in the second half of the previous century, there is a fraction of about 10-15% of cases who will fail primary therapy and who will be difficult to manage. Also to date it remains difficult to identify them up front.

So you need more refined tools and that’s exactly what you’ve been trying to do in this study.

Yes, that’s right.

What did you do?

The problem in understanding the biology of Hodgkin lymphoma is that the malignant cell of Hodgkin lymphoma, the Hodgkin Reed–Sternberg cell, is quite rare in Hodgkin lymphoma biopsies. It represents 0.1 – 1.0% of the cells in the biopsy.

As an aside, could you explain why that is?

That means that doing genomic or molecular studies on the malignant cell of Hodgkin lymphoma is extremely demanding. You cannot just take a biopsy or a piece of a biopsy and do studies on this tissue or on DNA or RNA extracted from this tissue. What you need is you have to purify these cells by laser microdissection and that’s very laborious and difficult to apply in a clinical context.

And the step forward that you made was what?

A little bit by accident or incidentally we found that by analysing circulating cell-free DNA from the plasma of patients with Hodgkin lymphoma allows us to detect and identify genomic imbalances present in the Hodgkin Reed–Sternberg cell.

Could you take me through what you did to discover this?

That was the discovery arose from an incidental observation. The point is that we recently introduced non-invasive prenatal testing in our department which is based on low pass sequencing of cell-free DNA. Then at some point we came across a young pregnant woman who had a massively abnormal profile so the profile did not point to any constitutional abnormality like Down’s Syndrome but we could not explain it by abnormalities in the foetus, nor in the patient. So at that point we asked the question could it be an acquired abnormality cancer which is causing this. This ended up or this led to a diagnosis of Hodgkin lymphoma. Next what we did is we started a small prospective trial in which we recruited nine patients with a new diagnosis of Hodgkin lymphoma and we looked at the cell-free DNA in the plasma and we looked by FISH at the Hodgkin Reed–Sternberg cells in the biopsy and we found a very good match between aberrations in the cell-free DNA profile and the result by FISH.

So you were able to find a signature for the existence of Reed–Sternberg cells because of the DNA that you found, the cell-free DNA?

Yes.

And how did you apply this, then, or how are you going to apply this to looking at these genetic imbalances and actually interpreting them in terms of things like prognosis?

Of course this is only a small study, we only did ten patients so far. So what we provided in our study is a technical proof of principle that by sampling cell-free DNA and analysing that from the plasma of patients with Hodgkin lymphoma we have a direct, or rather straightforward, manner of obtaining information on the Hodgkin Reed–Sternberg cell. First of all what we will have to do is to recruit more patients in the context of clinical trials but what I see is two important perspectives as of now. What we have seen in our cases is that the abnormalities in the profile disappeared rapidly upon early treatment, within one cycle or two cycles of chemotherapy the genomic representation profile had been restored to normal. So what could be an interesting possibility is that this could become a non-invasive tool to monitor interim disease diagnostics as PET CT is normally done now in clinical trials to look at the interim disease response.

And what about that important question of whether in the particular patient you have in front of you you could go for a gentler therapy to avoid the late toxicities.

This is also something which we have to develop and explore in clinical trials. But what we have seen in our initial series of ten patients is that we can see differences between patients, we can recognise some heterogeneity between patients. One of the important questions that is arising now is could this heterogeneity in patients help us to distinguish between those patients who will respond well to first line therapy or those who will not respond to first line therapy. It might also help in clinical trials, it might also help to find out which patients could benefit or to recognise the pattern of the patients which could benefit from therapy de-escalation.

So what should doctors be expecting to come out of this in the near future, then? If you could sum that up in a nutshell.

The first thing we have to do is… my idea would be to incorporate CCF DNA, a cell-free DNA, profiling in clinical trials looking at therapy de-escalation, for instance, and then what we will be able to do is to correlate the profiles with those patients who relapse, those patients who respond very well or maybe those patients who respond adequately to a de-escalated therapy. How I see this new possibility, I see this as a new research tool to look at Hodgkin lymphoma patients and to try and address a number of translational questions which remain unresolved now.

It looks like we have a novel gateway to the genome of the Hodgkin Reed–Sternberg cell and that is a novel gateway which is practical, which could be convenient and which could be also applicable on a large clinical scale in large clinical trials, much better than the current possibilities which require laser microdissection and other sophisticated techniques which are available only in a research context.