European Multidisciplinary Conference in Thoracic Oncology (EMCTO) 2013

Opportunities and complications of PET CT

Dr Barbara Fischer – Rigshospitalet, Copenhagen, Denmark

Great to have you with us, Dr Fischer, to talk about this important session on diagnostic work- up because PET-CT is being adopted but you are billing it as an opportunity and potential for pitfalls. What are the opportunities and pitfalls in using PET-CT?

The opportunity when using PET-CT for pulmonary nodules is that we can find out which nodules are benign and are better left alone and which are malignant which demand us to make further examination, invasive biopsies, etc.

So what have you been discovering in your work about how best to do this diagnostic work-up?

That is if you find a nodule, an unspecific nodule on a CT scan you can really tell whether this is benign or malignant then you can do a PET or a PET-CT which gives you an impression of the function of this nodule. When you do a PET you inject radioactive labelled glucose which is taken into tumours and into infections. So if it’s easily seen on PET then you know that something is going on here and we have to find out what is going on. On the other hand, if it’s completely negative on the PET scan without any uptake then you can safely say OK, we’ll come again in, for example, three months, six months, depending on other characteristics of the pulmonary nodules.

So just how sensitive and how specific is PET-CT for correctly identifying that a suspicious lesion really is a nodule containing cancer cells?

That depends on the patient, group of patients, where you test PET, that’s the problem. But in the early studies we found that it was very, very sensitive – above 90%, 90-100% sensitive. In numerous studies where we have included more patients with smaller, very small, nodules as well, the sensitivity decreases down to 70%. So that is definitely one of the pitfalls. You need to have an impression of what is the pre-test probability of cancer in this group of patients that you’re examining. The specificity of PET is around 70-80% so it’s not perfect and that’s mainly because you cannot see whether it’s a malignant nodule or whether it’s infection. So we have to work on the sensitivity and the specificity and that is where the field is going now. Sensitivity you can increase by, for example, doing respiratory gating where you follow the nodule with your PET acquisition, that would increase the FDG uptake in the nodule and would make the localisation, compared to the CT scan, more precise and make the procedure more sensitive. Also regarding specificity we are now beginning to test other tracers than FDG. FDG which is quite unspecific but we have another tracer, for example FLT, which is more specific for cancer. But this is quite new data and not in the clinic yet.

If you get your nodules correctly identified and they are cancer, what’s the benefit from that?

It’s important to identify a malignant nodule, a lung cancer, as early as possible because if you find it early on you can cure the patient by means of surgery or radiotherapy. If a long time passes the cancer will spread and the only treatment available will be, to a certain extent, radiotherapy or chemotherapy and that will often not be curative but only palliative treatment. So the earlier the better. On the other hand, we have now started to do screening trials using CT to screen patients, usually patients above 50 years of age and with a history of smoking, and we find a lot of small nodules in these patients. But only a really small minority of these nodules are actually malignant. So we can’t do a biopsy on all these patients, a biopsy has some limitations and could be potentially dangerous, so we need a second step test to find out which patients will benefit from biopsy and which patients will be better left alone or should go directly to surgery. And that’s where PET comes in.

From your perspective of using PET, then, could you describe to me the ideal diagnostic procedure that clinicians should be following?

You could say that if you have a solitary pulmonary nodule, it should be quite small, smaller than 3cm, probably also around 2cm and larger than 7mm say, and you cannot tell on CT alone whether it’s malignant or benign you should do a PET. That’s the short version. Then, depending on what you see in PET the next step will be different. If there is no uptake on PET you can safely tell the patient to come back for a follow-up CT after 3-6 months. If it looks malignant we will probably send this patient directly to a biopsy or to a surgical procedure. So that is the recommendation currently.

How does all of this mesh in with screening for high risk patients?

PET has potentially a very important role there because when you screen high risk patients you find a lot of benign nodules and you have to be sure how to handle these nodules so that you will not impose some kind of failure or some kind of unwanted thing during the biopsy in a patient who doesn’t have a cancer.

The burden of anxiety.

Exactly, also you have to relieve this burden of anxiety from the patient’s shoulder as quickly as possible and as safe as possible.

And how near are the refinements of PET which could increase that specificity to get it closer to 100%?

That’s a good question. It’s not right around the corner. I think what is right around the corner is that we use this wisely and we integrate information from CT and PET. Also there when you look at your PET scan you have a lot of information from the referring clinician so that you know how to interpret the PET-CT scan. The respiratory gating and the more specific tracers are coming but they are very laborious and the logistic is challenging so I cannot tell when, it’s not tomorrow, that’s for sure.

So one of the benefits of PET, let me get this straight, is that you can withhold treatment which is not needed?

Yes.

And the pitfall is that you might get it wrong but not too often at the moment.

You might get it wrong but when you are diagnosed, we’re talking diagnosis here, you have this pulmonary nodule and as PET is very sensitive the risk that you will miss a malignant tumour is very, very small especially if you use information from a follow-up CT scan as well. Then you will almost have a sensitivity around 100%. If you get it wrong the other way that you say it’s positive and it turns out that it’s, for example, an infection or sarcoidosis or something like that, then of course you didn’t find a malignant tumour in the patient but you found something else that might need attention as well. So it’s not that bad after all.

So how would you sum up the take home message for cancer doctors everywhere?

The take home message should be that PET-CT with FDG is very useful to discern between malignant and benign tumours but you have to use it carefully.

Barbara, thank you for giving us that knowledge for ecancer.tv.

You’re welcome, my pleasure. Thank you.