Primary mediastinal B-cell lymphoma: Radiotherapy superfluous post-chemoimmunotherapy

Share :
Published: 6 Jun 2023
Views: 115
Prof Emanuele Zucca - Oncology Institute of Southern Switzerland, Bellinzona, Switzerland

Prof Emanuele Zucca speaks to ecancer at ASCO 2023 about the results from the IELSG37 randomised trial which evaluated observation vs radiotherapy in primary mediastinal B-cell lymphoma patients with complete response to standard immunochemotherapy.

He explains that since the development of immunochemotherapy regimens, there has been an ongoing controversy over whether radiotherapy is still needed following immunochemotherapy in this setting.

Prof Zucca reports that patients without radiotherapy showed similar outcomes to those with radiotherapy.

Read the related news here.

The trial is about a particular type of non-Hodgkin lymphoma, primary mediastinal B-cell lymphoma, which is an aggressive entity affecting mainly young people, particularly women but not just them, in the ages between 30 and 40. This disease is characterised by usually a good outcome but poor outcomes for those who are not achieving a quick response to induction therapy or who relapse early after front-line immunochemotherapy.

The study was designed just to address whether we can avoid giving radiotherapy to all patients, or nearly all the patients, because this is associated with a high risk of secondary cancer and a long-term risk of cardiac disease, such as ischemic or valvular cardiac disease.

The point is that to avoid relapse, historically all patients after immunochemotherapy got mediastinal radiotherapy. This is a disease where you have a mediastinal radiological lesion in nearly all patients. Before PET scans it was impossible to understand whether this was residual cancer or just a sort of scare. For this reason, historically radiotherapy was given to these patients but about 15 years ago the NIH developed a novel chemoimmunotherapy regimen called dose-adjusted EPOCH which is mainly used in the US but not only. This regimen was promising enough to let many doctors no longer give radiotherapy to these patients.

There has been a never-ending controversy between those who believe radiotherapy was no longer needed and those who are still giving radiotherapy to everybody, such as most doctors in Germany and Italy, for example, whilst in France and the US there is a more balanced distribution. To end this controversy we decided to address this issue and randomise patients achieving a complete response according to PET scan between radiotherapy to the mediastinum versus observation alone. 

The study showed that patients without radiotherapy and those with radiotherapy had nearly identical outcomes. We had roughly half of the patients achieving a complete metabolic response according to the PET scan which was centrally re-assessed by a small panel of international experts. The randomised patients received radiotherapy or nothing and the actual benefit of radiotherapy in progression free survival is around 2%. But at three years after study entry the two arms have an identical overall survival of 99%. In other words, we should give radiotherapy to more than 100 patients to spare one single progression. This is clearly indicating that radiotherapy is safely omitted in this population.

How can these results impact the treatment of B-cell lymphoma?

This is going to change practice. There will be no more indication to give radiotherapy to patients in complete remission according to a PET scan. This will not only reduce the risk of long-term toxicity and morbidity associated with radiotherapy but also would reduce the overall cost of therapy.

There is still something that we will have to explore which is the outcome of patients in complete remission. In our study patients not achieving a complete remission but in partial remission have an excellent outcome if irradiated. But perhaps that contains a sizeable subset of patients who are actually already cured and may also be spared radiotherapy. This might be the topic of our next trial.