Oncotype DX Roll Call!

This is my first post. I have been struggling for last 2 months trying to figure out what my treatment plan will be.  This is my second dx with BC after a 10 yr cancer free period.  They are not sure if this is a recurrence or second primary.  Recurrence is thought because of close proximity to first site.  New Primiary is suspected because pathology is so different.

I just got Oncotype results back from surgeon and my score was 65.  I have no percentage score just the number.  How do you calculate the %? It seems like a score of 65 is pretty bad given the posts on this thread.

Any thoughts would be helpful.  I am going to an oncologist at Sloan Kettering for a second opinion as the first the first oncologist said I should see only a BC specialist.

2xBC -- We're twins.  I was diagnosed in May 08 with IDC, 7mm (<1cm), ER/PR/HER2 positive, Grade 3, negative nodes, negative margins.  My oncotype score was also 65.  To date, other than hearing about you, I haven't seen anyone scoring in the sixties.  My oncologist, who has been practicing for 27 years, said that he's never seen a score that high (keep in mind though that oncotype testing in only a few years old).  Also, as a rule, HER2+ people don't usually get the oncotype test done, because positive HER2 status alone, merits chemo.

 Now, to understand the 65 score...

With scores from 1 through 30, you're risk of recurrance rises with the score, i.e. a 10 score might show a 7% chance of distant recurrance....a 25 score might show a 15% chance of distant recurrance, etc., etc,.  ALL scores higher than 30 return a recurrance score of 35%.  So...whether your score is 31 or 99....you're recurrance statistic is 35%.  The reason for this is because (now THIS is scary....), there were so few people with scores over 30 that they never had a statisically significant sample size associate with each score over 30 to predict real recurrance percentages.  Because of this, Genome (the makers of the oncotype test), took the AVERAGE recurrance rate of all scores over 30.

 So, in plain english, in people like you and I, with scores in the mid-sixties, our recurrance score  of 35% is probably fairly accurate, since 65 falls midway between 30 and 100 (almost anyway).  According to my oncologist, chemo would reduce that by around 11%, bringing the recurrance rate down to 24%.  Then, adding Herceptin, divide the 24% by 2....bringing the absolute risk to around 12%.

In a nutshell, you're headed for some aggressive chemo.  I, like everyone else here, was hoping to dodge that bullet, but no such luck.  It's over now (the chemo and radiation), and aside from not recognizing myself in the mirror with this atrocious hair-do, I'm really glad I did it.  Best of luck to you and I'm so sorry that you're having to go through this a second time.

Kim

Seabee, it's the graph on page 1 of the Oncotype report that is used to calculate the 10-year risk of recurrence based on the Oncotype score.  I think we need to forget about the graphs on page 2 as far as calculating recurrence risk.

There is a section halfway down page 1 of the Oncotype report that is labeled "Results".  It shows the Oncotype "Recurrence Score" (which we've been calling the "Oncotype score").

The section beneath the "Results" section is labeled "Clinical Experience: Prognosis for Node Negative, ER-Positive Patients."  The text in that section says, "The Clinical Validation study included female patients with Stage I or II, Node Negative, ER-Positive breast cancer treated with 5 years of tamoxifen....".  That section contains a graph which is  the standard curve used by Genomic Health to convert Oncotype "Recurrence Scores" into the (%) risk of recurrence.  See the label on the vertical axis of the graph?:  "Average Rate of Distant Recurrence at 10 Years after 5 Years of Tamoxifen Treatment."

That means the graph used by Genomic Health to convert someone's Oncotype score into a 10-year risk of distant recurrence is based on results after 5 years of tamoxifen.  The graphs on page 2 of the report (the "Chemotherapy Benefit" graphs) weren't even included in the Oncotype report I received back in March of this year.  I don't know what those graphs are good for, since the study numbers are so small.  I do know that Genomic Health does not calculate or report recurrence risk from the graphs on page 2.

2xBC, Kim is right--there weren't enough women with scores as high as yours to provide valid numbers for the Oncotype standard curve.  The standard curve Genomic Health was using this past year (and the one on their website) cluster and average the recurrence risk for scores above 50, not above 30.  They have enough data to report individual Oncotype results for scores between 30 and 50.  OTOH, as you've heard from your onco, a tumor that is ER- PR- HER2+ probably won't generate an Oncotype score with a reliable recurrence risk, because the assay wasn't tested on ER- tumors.

If your tumor is ER- PR-, and especially if it is HER2- but even if it's weakly HER2+, you will very likely be getting chemo.  Hormonal treatment isn't effective on ER- tumors.

Moody ... sheesh.  I don't know what to say.  AdjuvantOnline is a traditional way to calculate recurrence risk.  I read somewhere that it's used by 75% of oncos.  The results aren't the same as for Oncotype testing, though, because AdjuvantOnline doesn't take HER2 or PR into account and the chemo database is out-of-date.  AdjuvantOnline used to have a section that took Oncotype scores into account, but that's been taken off-line.  There's a newer version (v. 9.0?) that has supposedly been in the works for the past year.  No sign of it, yet.

otter 

Otter--Yes, I should have read the fine print.  But since a poor excuse is better than none, all I can say is that on my fuzzy copy of the report, not all of the fine print is legible.  I now see that the graph on p. 2 is based on a different patient sample, although they have the same clinical features (node negative, ER+, etc.) as those the first graph represents. That could account for some differences in numbers, but overall the curves look very similar.  My report has a third graph presenting curves for node positives. This graph is based on an even smaller sample of 367 and spans only five years.  However, the percentage of risk remains the same for my recurrence score of 16 on all three graphs.

Unfortunately, my game attempt to make sense out of Moody's numbers has failed. But I notice that Kim has been flashed some numbers that sound somewhat similar--an 11% reduction with chemo, and then the 24% remaining risk cut in half by Herceptin. Where does this stuff come from? An oncologist told me that out of a hundred women with my clinical profile, 56 would survive without further treatment after surgery, 32 would die of cancer, and l2 would die of other causes. Hormonal therapy would reduce the number who died of cancer by 9%--that is, only 23 would die of cancer. Chemo would reduce this mortality by 11%, so that only 21 would die of cancer. And with both hormonal and chemo, only 15 would die of cancer, reducing cancer mortality by 17%. This left we wondering, if chemo saved eleven lives and hormonal 9, why mortality wasn't reduced to12. It also left me wondering how anyone had been able to isolate a group of 100 or more women with exactly my clinical profile and follow them through x number of years in order to record these results. In short, I was skeptical, and decided to work with another oncologist who was less preoccupied with mortality.

P.S. I now know where my onco #1 got those figures. They are rounded off from the figures provided by Adjuvant Online for someone of my age, tumor size, and # of nodes involved. This is preceded by a lot of waffling about the controversies surrounding adjuvant therapy, especially chemo, and a warning that the projections are hypothetical. No kidding!  In the supplemental windows, one finds more disclaimers and hedged bets, as well as as lot of citation of *relative* percent of improvement from first to second to third generation regimens. They often seem to be saying, "Chemo seems to be improving slightly, but we're not sure of that, or of much of anything else." I can hardly wait for the updated genomic model.

Seabee, I'm beginning to think Moody never had an Oncotype DX test done.  (Are you reading this, Moody?)

I went back to the original thread she started, and looked at her original post.  This is what it said:

"I keep reading about peoples oncotypeDX score.  My onco didnt give me a score, he gave me a percentage.  After surgery and no other treatment I had  75% survival/25% recurrence.  And after my treatment options I could lower to 92% survival/8% recurrence.

How do I know what my oncotypeDX score is?"

That sounds like "pure" Adjuvant!Online prognostication.  Each treatment option produces a change in the recurrence risk.  The changes are not additive.   There's nothing in there about Oncotype testing.  Moody, have you asked anyone about that, yet?  It's a moot point, at this point, since you've already had chemo; but it would explain the difficulty I've had understanding your percentages.  If you never had an Oncotype test done, I shouldn't have used your 25% recurrence risk to back-calculate an Oncotype score.

Also, Moody, when you subtract the [% recurrence] from 100, the result isn't the "% survival."   It's the "% recurrence-free."  One might assume those women survived, but that's not necessarily true.  They could have had a recurrence later.  Also just because someone has a recurrence, doesn't mean she isn't in the "survivor" group.  Recurrences can be treated, and someone with a recurrence can be in remission ("no evidence of disease") for years afterward.

[I'm in a trouble-making mood today, I guess.]

otter 

Wow, Kathy, I don't think I've ever seen such a nicely written set of guidelines for HER2 testing and test interpretation.  Thanks!  (And, you're right about the way the HER2 results are presented in the Oncotype DX report.)

Of course, what all this says is there is disagreement among all 3 of the tests, and nobody knows for sure what it all means.  According to the document in the link you gave us, there is no true "gold standard" test for HER2 over-expression.  Some oncos are erring on the side of over-treatment and are giving Herceptin even to women who have "equivocal" HER2 results.

otter 

I stumbled onto this thread, and it's the best explanation of the oncogene dx I've seen: I finally got the third page from my surgeon--they kept faxing me the first two pages, which do fax so poorly that you can barely read the recurrence score.

I'm recurrence score=12, 8% risk, I was advised to just take tamoxifen, which I've been on since August.

Due to a lot of factors, I've lost some confidence in my treatment team, and am going for a second opinion at Dana Farber this week--7 months after surgery, but I'm the one who had to tell the oncologist that one of the medications I take interferes with tamoxifen... 

Otter, you never cease to amaze me.

Kira