Levo and cancer risk - interpretation of stats

[Eton]

For info following recent discussion on this subject. I am not a statistician so the figures in this paper are fairly meaningless to me. Perhaps someone with a statistical background could give a layman's interpretation for the benefit of other?

onlinelibrary.wiley.com/doi...

[jimh111]

I'm writing up a topic on this for my web site. It will be about a month or two before it is ready. I have to finish writing it and then check what I have written is reasonable! It's a very sensitive topic so I want to keep it level headed. I'd prefer not to answer lots of questions as I need to get on with it.

As regards this paper. First it looks at cancer risk - how likely someone on levothyroxine is likely to get cancer compared to the general population. This is a good study. It's important that such studies look at subjects a few years before they get cancer (this one does). The reason is that cancer can play havoc with thyroid hormone levels. A downside of such studies is that they tend to take a single snapshot of e.g. TSH, fT3, fT4 and so it is possible some people will have different levels later on, especially if they are treated for abnormal thyroid hormone levels.

This study looks at the risk of getting cancer, others investigate how many die from cancer.

Some cancers differ in their response to thyroid hormone. Liver cancer tends to fall with higher hormone levels, perhaps because lipid levels are lower. Breast (and possibly prostate) cancers tend to increase with higher thyroid hormone due to higher sex hormone levels.

The easiest way to understand this study is to look at 'TABLE 2', specifically the last two columns 'Adjusted odds ratio' and 'P value'. First only look at those lines (rows in computer speak) with a P value of 0.05 or less, results for higher P values may have arisen by chance.

You can see e.g. the adjusted odds ratio for all cancers ('overall') is 1.50 with a P value of <0.0001. This tells us that patients on levothyroxine are 50% more likely to get cancer than the general population. The P value tells us that there is a high degree of certainty for this conclusion.

The situation gets more complex because we should exclude breast cancer from the overall figure because the increased cases may be due to patients having higher fT3, fT4 than average (they need a higher fT4 to get an average fT3). Another potential confounder is that hypothyroidism prior to diagnosis may cause a higher cancer risk - I don't think it does but these sort of things are possible.

I've seen other studies which suggest a higher overall cancer risk of about 10% (odds ratio 1.10) for those taking levothyroxine. So, it does seem that people on levothyroxine have a higher risk of getting cancer. This subject is complex but very important.

[TSH110 in reply to jimh111]

Fascinating I wonder if this is also so with NDT. There is a lot of non Hodkins lymphoma in my family and a lot of thyroid disorder (all in the same branch of the family) so perhaps they are related. Your paper sounds very interesting will look forward to reading it.

[Eton in reply to jimh111]

Thank you Jim for your reply which has helped me understand the stats a little better. It is so worrying to think that those of us taking levo are 50% more likely to get cancer than the general population. That's a huge difference between those taking levo and the rest of the population.

It would be interesting to know if those with naturally high levels of Free T4 (ie not taking replacement hormones ) are similarly at risk. Also does same apply to those taking Lio? Does length of time on hormone therapy play into the situation? So many questions which are impossible to answer.

In my circle of friends/ acquaintances I know of 4 women taking levo who also developed breast cancer. However I also know others who got the disease and were not on levo. I also have a friend who died in early 50s from pancreatic cancer and was taking levo .

The only thing we can do is to try and live as healthy a life as possible and seek medical advice if we are concerned. Early detection of cancer improves outcomes etctec

I look forward to reading your paper in due course and do not wish to distract you any further.

Many thanks again.

[jimh111 in reply to Eton]

These are the sort of questions I’m trying to raise. There’s quite a bit of recent research that suggests it is T4, natural or from tablets that has a unique role as well as overall hormone levels. I want to raise the issues and leave it to the oncologists to handle the detail.

[thegemprincess in reply to jimh111]

jimh111 Hi Jim: Are you studying people who take compounded formula? I see they may have a greater chance of developing breast cancer. Is that because the formula to treat hypothyrodism raises Estrogen? I am 64 and out of nowhere when I had a slight increase in compounded formula have hot flashes. Is this possibly linked to increase in thyroid meds? Your response is greatly appreciated.

[jimh111 in reply to thegemprincess]

First, I have no expertise in cancer, I'm looking at it as a thyroid patient. The studies I've seen suggest breast cancer is a little different to other types as regard thyroid hormone. It seems that thyroid hormones increase breast cancer risk by increasing sex hormones as you have observed. This applies to all forms of thyroid hormone, the higher the hormone level the higher the risk. I appreciate some patients need higher thyroid levels so the best I can suggest is that they are extra vigilant.

[tattybogle]

Thanks for posting this .. good find .

I just had a quick look .... it's a bit complex , but then it would have to be cos it's a complex subject.

On the plus side ..... I think ? they found Levo use was associated with LESS cases of cervical cancer,

and that for most (but not all) of the types of cancers they looked at , they did NOT find more cases of the cancer associated with the higher doses of Levo . I was fearing the data would find higher dose was associated with more cases .. but it seems not . or mostly notSo we must be careful not to jump the gun by assuming ' a cancer' means 'all cancers'

the "50% more " did cause me to have a bit of a sharp intake of breath, but as they say in their conclusion , they have far too little information yet to know why they are seeing this association , and association is not causation.

So i think for me it's still a case of "Keep Calm and Carry On " ( and keep my eyes and ears open)

[HashiFedUp]

I think there is more research needed here and I am not sure if they had enough biodata on participants. ie lifestyle stuff which we know is the biggest cause of cancer. eg weight diet smoking etc. I also think that thyroid patients have over active immune systems and are statistically more likely to develop other auto immune diseases, of which cancer is classified. So more research needed and if we all try and stay as healthy as we can through lifestyle, that hopefully will help!!?