Proton Therapy Congress 2016

Health outcomes and QOL in paediatric proton therapy 

Dr Torunn Yock - Massachusetts General Hospital Cancer Center, Boston, USA

I’m talking about the role of proton radiotherapy in treating paediatric cancers. I’m focussing on primarily the paediatric brain tumour population because we have the best clinical data that is quantitative in this cohort. The primary message is that we are starting to make real inroads into proving the benefits of proton radiotherapy in the paediatric population. At Mass. General we’ve studied the neuro-cognitive outcomes in our paediatric medulloblastoma cohort and we published this in Lancet Oncology earlier this year. These children get a component of craniospinal radiation which has a component of whole brain radiotherapy and then it’s followed by an involved field boost so we do expect that there will be some fall off in neuro-cognition. But the fall off isn’t quite as steep as it is in the photon populations and during my talk I go over the data in our proton population as well as the photon population and where the pitfalls in interpreting the data are. There are a couple of key points to remember when you’re looking at outcome data other than disease free survival so we’re very good at listing the follow up time point for disease free survival - the median follow up is five years or seven years - but we are bad at listing the median follow up for other outcome measures and yet this is absolutely critical. The longer you follow a patient the more likely you’re going to see an adverse outcome. So when reviewing the literature the median follow up for that outcome is critical.

So when looking at the neuro-cognitive data it’s very important to note what the median age of the group is as well as the follow up and what metric they’re using. There are a number of pitfalls that can lead you astray with regards to neuro-psych outcome. So median age, age determines the neuro-psych outcome almost over anything, so age is most important. Secondly, the metric used: I reported full scale IQ which measures processing speed which is the component of full scale IQ that is most affected. A lot of current studies are now using estimated IQ which is a very shortened version which has no measure of processing speed so it artificially inflates the measure of IQ and really should not be used in the brain tumour cohort as it’s a poor measure for research and it doesn’t serve the patient well either.

You also mentioned hearing loss as another potential outcome, could you elaborate on this?

We reported our data with regards to hearing loss in a medulloblastoma cohort and basically what we found is approximately a 9% benefit in reduction of the grade 3 or 4 hearing toxicity rates. So we also go over our neuroendocrine data and show that the rate for neuroendocrine deficiency of five years in the medulloblastoma cohort is about 55% but when you look in the context of other literature, this is looking at photons, it’s within a range. So what we’ve done is paired with Emory to compare our neuroendocrine outcome compared to their endocrine outcome. What we’ve found is that in the standard risk medulloblastoma cohort, and we took just the standard risk so we could standardise the dose of craniospinal radiation, we found that there is a benefit for the patients treated with protons with regards to thyroid deficiency, overall hormone replacement risk and then actually height as well as sex hormone deficiency. The height is probably not real, that is probably related to the fact that when we find a growth hormone deficiency we’re very good or aggressive about replacing the growth hormone; other institutions have different theories about the benefits of growth hormone replacement so it’s not always done.

Could you tell us more about quality of life?

We have embarked on a comparative study with Stanford University cohort of paediatric brain tumour patients to compare quality of life outcome as measured by the previously standardised survey of the paeds’ QOL. What we found is that in overall our paediatric brain tumour population appears to score higher by ten points compared to the Stanford population. That is statistically significant, it is a very suggestive study, that there’s a quality of life benefit to protons but it is not without some caveats. So it’s not definitive but suggestive; there are some limitations to the study in that we do not account for socioeconomic status and we measured the proton patients more recently than we measured the photon patients and error can make a difference as well. But we are presenting comparative data in a medulloblastoma cohort from Mass. General and compared to the Royal Marsden medulloblastoma cohort at SIOP. I can’t give you the details but we will present it at SIOP in Dublin this year, which is in October.