I left the San Antonio Breast Cancer Symposium Saturday (yesterday, as I write). I had been there since the previous Sunday, and was all meeting’d out and ready to go home. The morning was grey and misty, almost gloomy. As the plane broke through the clouds, we were suddenly bathed in an almost blinding light.
When you write you are always searching for metaphors, and when one literally strikes in you in the face, you accept it for the gift it is. Cancer researchers often toil in a grey mist, a world of indistinct outlines, an oft-depressing, or at least frustrating, place. Every now and then things come into focus, lit by the brilliant and clarifying light of science. I felt that way flying home from San Antonio.
I have, I must admit, a foolishly sentimental attachment to San Antonio. I trained there with Bill McGuire, the symposium’s founder, and with Kent Osborne, its current director. My eldest son was born there. I developed a love for Tex-Mex food and for the pleasant rhythms of the city. The meeting itself allows me to conduct a great deal of business, and to catch up with old friends.
Some years that opportunity to re-visit old haunts and old friends is all I get out of the meeting. But some years I come home re-charged, and delighted by what I have learned. This was one of the good years. The meeting brought into focus both what we have accomplished, and where we may be going.
Like most breast cancer researchers, my cognitive map of the disease revolves around biologic subtypes, in particular the estrogen receptor, HER2, and so-called triple negative disease. There was progress for all three this year, and some interesting side-trips as well.
HER2-positive breast cancer was the area of clearest definition. Adjuvant therapy for HER2-positive disease dates to the stunning 2005 ASCO Annual Meeting, one of the great moments in the history of our profession, with the presentation of the Joint Analysis of the NSABP and NCCTG trials and of the HERA trial.
Both the Joint Analysis and HERA were updated at San Antonio. Both continue to show clear, impressive, and long-lasting benefits for adjuvant trastuzumab. This year’s Joint Analysis, we were told, was the final one for this epochal study. I mourn its passing, even as I celebrate its success. I remembered, in listening to Ed Romond’s presentation, Winston Churchill’s description of a great World War II battle: if it was not the beginning of the end, it was at least the end of the beginning.
Both HERA and another trial, PHARE, examined the question of trastuzumab duration, and both point to a year of adjuvant trastuzumab as the “sweet spot.” Other large trials are still examining the duration question, but for the moment we have a working answer. And it is the same answer as last year’s answer.
The HER2 universe expanded slightly at San Antonio. An analysis of HER2 mutations performed by a very good genomics group at Washington University suggested that somewhere in the range of 1.5-2% of breast cancer patients have activating mutations, usually in what we think of as “HER2-negative” tumors, cancers with neither HER2 amplification (measurable by FISH) nor overexpression (measurable by immunohistochemistry). Lovely preclinical work suggests that these cancers might be sensitive to the small molecule kinase inhibitor neratinib, currently in clinical trials for HER2-positive breast cancer.
I found this study interesting for what it points to in our future. If one has to screen 50 patients to find one with a HER2 mutation, then one has to screen 500 patients to perform a 10-patient pilot therapeutic trial. We will see this again and again in the genomic era, where rare driver mutations are routine, and it will force us to change the way we study the disease. A consortium of investigators is now actively screening patients for HER2-activating mutations for entry into a neratinib trial.
Hormonal therapy also saw evidence of forward progress at San Antonio. Two forward steps involved old drugs: tamoxifen and fulvestrant. The ATLAS trial (Adjuvant Tamoxifen, Long Against Short) finally gave us something to sink our teeth into regarding tamoxifen duration, showing that 10 years was superior to five with regard to overall survival. The difference was not huge (about 3 women in a hundred), but ATLAS joined MA-17 in suggesting that longer duration hormonal therapy (SERMàSERM or SERMàAI) is superior to shorter duration. We’ll now need to see whether longer duration aromatase inhibitor therapy also wins out in the MA17-R trial.
I was quite cynical about fulvestrant in the past. It didn’t seem to add much above and beyond older hormonal therapies. But recent years demonstrated that we quite simply got the dose wrong, and that 500 mg (administered monthly as an intramuscular injection) was superior to the original 250 mg dosing. The original evidence of superiority involved progression-free survival, but this year’s San Antonio extended this to a demonstrable improvement in overall survival, with little added toxicity (other than an additional sore buttock every month).
The other big news in the estrogen receptor space came out of left field, one of those unexpected delights I didn’t see coming. A randomized Phase II trial of letrozole with or without the cyclin dependent kinase 4/6 inhibitor PD0332991 showed a tripling of progression-free survival, and minimal “paper” toxicity (some neutropenia). The clinical work was based on preclinical studies suggesting a synergistic interaction between estrogen receptor blockade and CDK 5/6 inhibition.
Like many old drug developers I am suspicious of randomized Phase II trials. The phrase I heard at least a dozen times at the meeting was “Remember iniparib” (iniparib being the Phase II rose that withered when a Phase III trial was performed). But the CDK 4/6 inhibitor was certainly promising: I’d be happy to put patients on the Phase III trial.
All in all, 2012 was a good year for hormonal therapy in breast cancer: the approval of everolimus in combination with exemestane, as well as the Symposium’s findings mentioned above. We are teaching the old dog new tricks.
Triple negative breast cancer (TNBC) remains a tough nut to crack. The genomics suggest a collection of orphan diseases, lots of mutations, and no kingpin a la HER2 or ER; instead, many little mutinies add up to a large and dangerous rebellion. But even here there was hope. A genomic analysis conducted by the Vanderbilt group suggested the presence of actionable mutations in patients at high risk for recurrence after neoadjuvant therapy. Kap Bhalla presented work suggesting that HDAC inhibitors might increase the chemosensitvity of TNBC by inducing “BRCAness”, and Andrei Goga and his colleagues described a novel form of synthetic lethality that might allow a novel attack on these cancers via PIM1 kinase. None of these is an actual clinical result, but all point the way forward.
Science magazine identifies a “molecule of the year” every year. If I had to do this for San Antonio, I would choose PI3K, in large part for its Zelig-like ability to show up in every tumor type, and take on a different personality in each case.
In ER-positive breast cancer, PIK3CA mutations are common (relatively speaking) and are associated with better outcome, as an analysis of the TEAM trial showed. In contrast, in HER2-positive disease PIK3CA mutations are associated with resistance to HER2-targeted therapy, as a lovely biomarker analysis of the CLEOPATRA trial demonstrated. In triple-negative breast cancer, where mutations are rare but pathway activation is common, PI3K may be a useful therapeutic target when combined with EGFR/HER3 inhibition. PI3K was the great crosscutting story at this year’s meeting.
A few other presentations caught my attention. The CALOR trial randomized patients to receive or not receive combination chemotherapy after therapy for local-regional recurrence. This was a tough trial to complete: despite the involvement of three large cooperative groups, accrual came nowhere near its planned goal. Despite this, there was a statistically significant survival advantage for patients receiving therapy. Micrometastatic disease is still potentially curable after the failure of an initial attempt at cure.
The last decade saw several new chemotherapy agents enter the arena of metastatic breast cancer, the most recent being eribulin, a microtubule inhibitor originally derived from sea sponges. This year’s meeting saw a trial comparing eribulin and capecitabine in the anthracycline and taxane pre-treated metastatic disease. Eribulin won, just, with a (not quite statistically significant) six-week improvement in overall survival. No one looking at this trial (with the exception of optimistic marketers) would consider this a great triumph, for neither drug was very good. I viewed this trial as demonstrating the limits of cytotoxic chemotherapy.
Let’s conclude with a word of caution regarding neoadjuvant chemotherapy. Cortazar and an international consortium of investigators reported the effect of pathologic complete response (pCR) in the neoadjuvant setting in a meta-analysis of 12 randomized trials. As with reports from previous trials, a pCR is correlated with better individual patient survival. But at a trial level, pCR is not a predictor of improved event-free survival.
Neoadjuvant trials have become a popular way of examining new agents and regimens, and the FDA only recently released new guidelines for drug approval based on pCR. The meta-analysis lands like a large, cold, wet blanket on both the FDA guidelines and the belief that we can use neoadjuvant therapy to choose winners (a belief that spawned innovative approaches such as the ISPY2 trials). This disappointing result cautions us (if we really needed cautioning) that the stairway to therapeutic heaven in clinical trials is a long and hard path, with few shortcuts.
So that’s the news from the 2012 San Antonio Breast Cancer Symposium: a fascinating year for breast cancer geeks, and a good one for breast cancer patients. And, as usual, the margaritas and fajitas were great.