do 2 year disease free survival stats predict 5 yr disease free survival stats in
 

adjuvant breast cancer treatment studies? The answer is...sort of, but again they lumped all the different studies together instead of looking at premenopausal vs postmenopausal, ER+ vs ER- and her2+ vs her2-

Ann Oncol. 2007 Nov 20; [Epub ahead of print] Links
Correlation of changes between 2-year disease-free survival and 5-year overall survival in adjuvant breast cancer trials from 1966 to 2006.

Ng R, Pond GR, Tang PA, Macintosh PW, Siu LL, Chen EX.
Department of Medical Oncology and Hematology, Princess Margaret Hospital, Faculty of Medicine, University of Toronto, Toronto.
BACKGROUND: Although disease-free survival (DFS) is accepted as a valid end point in adjuvant breast cancer trials, improvement in 2-year DFS has never been formally established as an adequate correlate for 5-year overall survival (OS). We set out to ascertain if changes in 2-year DFS can be used to accurately predict 5-year OS changes. Design: We conducted a systematic Medline search (1966-2006) for randomized adjuvant breast cancer trials of >100 patients per arm with 2-year DFS and 5-year OS data. A univariate regression model weighted by trial sample size was constructed to determine whether 2-year DFS differences between treatment arms within trials were predictive of 5-year OS differences. RESULTS: A total of 126 studies containing 149 treatment comparisons met the inclusion criteria. Difference in 2-year DFS was a significant predictor of difference in 5-year OS. For every 1% increase in 2-year DFS difference, the 5-year OS difference increased by 0.5%-0.55%. The proportion of variation explained ranged from 0.38 to 0.42, with a wide prediction interval. CONCLUSION: There is a statistically significant correlation, of moderate strength, between difference in 2-year DFS between treatment comparisons and difference in 5-year OS but the correlation is not strong enough to be used as a predictor.
PMID: 18029973 [PubMed - as supplied by publisher]

Lani,

If I read what you posted correctly, this study is correlating 2 year disease free survival with 5 year overall survival, not 5 year disease free survival.

Hopeful

You are right--I mistitled it
 

of course, OS includes death from other causes as well as deaths caused by the treatment as well as the disease!

"The proportion of variation explained ranged from 0.38 to 0.42, with a wide prediction interval."

My actual background is in statistics and finance. I hated economic analysis studies when they tried to make something fit when it obviously did not. This is a prime example of the same thing in medicine...

What they are describing is R squared - roughly defined as the proportion of the square of the variance actually described by the model... In this case, their model is able to account for less than half of the square of the variance observed. This is a weak prediction in any field of study.

The term wide variance interval is also a danger signal. Think of it this way... If you had two classes, and they each had 5 students. The students in class 1 scored 80, 90, 85, 82, 88 on a test. Their average is 85. The students in class 2 scored 70, 90, 95, 70, 100. Their average is also 85.

Even though both classes averaged the same, the variance between the two is much different. In class 1, nobody was more than 5 points from the average. In class 2 there were multiple people up to 15 points from the average.

My Conclusion: The study can only account for approximately 40% of the square of the variation (weak predictor) and the prediction interval is wide - similar to the second classroom in my example above.

I also find it interesting that they did not include the level of significance used to come to their conclusion that 2 year DFS was a "significant" predictor of 5 yr survival. With such a weak model, and with their use of the wording "moderate strength" I would bet that they used .90 as their cutoff - leaving at least .10 that the results are just random chance. I would expect at least a .95 significance test be required before getting published, but maybe the Journal was desperate for submissions.

CPA,

Thank you SO much for the explanation! I get utterly bamboozled by much of the math that is thrown around in these studies. Your perspective is very helpful. Please, post about these numbers more often - you obviously see details lost to many of us.

Hopeful

I was disappointed that they, like so many STILL try to
 

lump all breast cancers together in order to reach out (very very far) to try to find these supposedly significant correlations.

How about looking only at studies of triple negatives, her2+s or even her2+er+s vs her2+ser-s, premenopausal vs postmenopausal, ibc vs ...

Statistical (more than far) reaching aside, its like asking if all human beings who live to age 2 also live to age 10. The difference is vastly different between those in areas of Africa where starvation and infectious disease are rampant vs Scandinavia vs Eskimos in the arctic.

How about comparing apples with apples?

CPA, So nicely explained. Thank you!

I really don't put much faith into any of the statistics indicated on these kinds of studies. I think it also depends on the perspective who wrote the findings and their knowledge of statistical sampling. This study just has too many different diagnostic variables and criteria And results that are only representative or that only predict approximately 40% accuracy are too weak and worthless in my opinion. The study conclusion indicates '...the correlation is not strong enough to be used as a predictor.' Why bother reporting or posting this study?

Yes, CPA do post more often. By the time these studies are offered to laymen, they are reduced to 3 words, without careful analysis of the numbers. I agree with Lani, this info would be more usefull if they broke it up by subtype. I guess the problem is they couldn't go back to 66 with various subtypes. Thanks, Bev

I bothered posting the study
 

as many times on this board I have heard the question asked--how do we know whether herceptin when used in the adjuvant setting just delays the onset of metastasis...how will things look five and more years down the road...when can one stop worrying

The answer is not known, but this study at least proposed to look into the matter--I had never seen any such study before-- although the study was poorly executed, both statistically and by lumping all breast cancer patients together.

Perhaps her2 has only been measured routinely (and only in some centers in some countries) for a few years, but the study could also have been divided by pre vs postmenopausal, ER+ vs ER- etc.

I particularly dislike studies performed by doing Medline searches of clinical trials and them lumping them all together. Not just comparing apples with oranges, but trying to make sense, or at least a point, with fruit salad (and a very varied one at that)

Lani, that would be a different study... ?????
 

When Herceptin is used in the adjuvant setting, does Herceptin delay or prevent the onset of metastasis over n periods of time? (criteria separated by type of bc, pre vs postmenopausal, ER+ vs ER-, bc stages, etc),

OMG, wouldn't knowing that answer be so grand !

With your network/contacts, is it possible you could find out if any researchers are working on this, or could you possibly suggest Genentech agree to responsibility for coordinating this, or could you offer the suggestion to any other potential researchers?

We HER2ers could even participate in the study/survey.

Joanne S

I am anxious to find out about the results of HERA III trail related to 1 vs 2 year administeration of Herceptin.

Joanne I think what you have posted may be oversimplified
 

there are supposed to be some HERA results at SABCS in December vis a vis
1 yr vs 2 yrs.

Whether they will have divided those out by ER+ vs ER- I don't know (I have heard there were very few "events"--a good thing--which is why they didn't think they could report this until later in 2008. That story changed when they announced they would give an interim report at SABCS.

Don't know if Genentech is really as interested as they were prior to getting Herceptin approved for adjuvant use. I think they are busy looking for other diseases/indications for herceptin use to broaden its market rather than looking back to find out whether survival was prolonged but metastasis occured, merely later (if so, it would not help their marketing). Sorry to sound jaded, but that seems to be the way things go.

Will certainly post if I find any other articles/news items/conference talks on the topic!

probabilities? ratios? volumes? differences? statistics? percentages? forecasting
 

Lani, You're obviously very intelligent and knowledgeable person. I don't know if you have first hand experience as a bc patient or supporter or how you've come to be involved in contributing to HER2. Perhaps it's a big secret. Funny, seems to me ---like--- you see me, but I can't see you. In addition, I can't even send you a private message or an e-mail? and there's nothing on your profile? I mean no disrespect---you probably have your reasons---it just makes me feel a little uncomfortable---like you are staring at my naked core (cuz I spill my guts out here), but I don't see you or know who you are or anything about you at all---other than you are someone who regularly cuts and pastes a lot of current study stuff. Many HER2ers certainly appreciate your posts.

I do try to make all these issues and concerns less complicated that's for sure. I'm not sure what's oversimplied, but I guess it doesn't matter at this point. All I know is that Herceptin does help a whole lot of HER2 people, but not all HER2ers.

So it sounds like it's not worth the effort to inquire or investigate or even discuss this any further. Sounds like it's all about the monies $$$$$---- so we won't know the answer to that recurrence/survival question.---that is unless I have a recurrence (still then I probably won't know if the recurrence it is a repercussion of Herceptin or not or something else and/or somemore something elses) or unless I die ---well then it won't matter to me anyway. But if I don't have a recurrence, I'll just be very happy and attribute it to all my personal diagnosis criteria and treatment combinations.

I guess that's the way I feel about most of these study results---probabilities? ratios? volumes? differences? statistics? percentages? forecasting? predications? inconclusive? additional studies required? where do I fall??? where do my her2 sisters fall? hum? I still have no clue after reading them.

I'm guess I'm pretty simple minded and I think I gotten myself totally wacko from not having any sleep for a couple of days and having anxiety as I will be leaving to the hospital for surgery in an hour.

Joanne-

I hope your surgery went well and that you are resting comfortably!

And I agree, the biggest shock (other than getting burned by stats in the first place being dx at 38 w/ no family hx) is that one bc does not fit all. How impossible it is to find the real truth for each of us and how we have to hope someone is looking!

Lani - I am very glad you post here. I know you have shared some information on your personal motivation and interests, over time. And, with the volume of information your provide us, were I in your shoes I wouldn't publish my pm or email either - it would be clogged just by me with the number of times I have personally wanted to ask you a question about the info you provide! I am glad you are here, we need people like you who are able to contribute, and I'll take your input and thank you for the research you do for us, on your very modest terms, any day!

Most of the hard questions I throw at my onc, come from inspiration I have gotten through your posts. Sometimes you post material even my onc hasn't seen - he is a young and bright man, and when I take him a printout he immediately looks it up on his computer (during my consult) and ALWAYS takes a grateful - this can only be helpful for me and all the women with bc that he treats. How can oncs possibly keep across all of the latest research for all of the types of cancer they treat?

Information is power Lani, and you help make us very powerful. Thankyou.

Bx

Lani, thanks for the post. And CPA thanks for the additional explanation.

In this case as you mentioned the correlation is weak.

I think the best we can say at the moment is that between years 2 and 5, the slope is less steep for the groups of women who took Herceptin in the adjuvant setting. But only time (and better designed studies) will tell when the slope starts leveling off.