THE CUTTING EDGE
EVIDENCE BASED INFORMATION
DIRECTOR, MEDICAL RESEARCH
LATE BREAKING NEWS FROM THE CUTTING EDGE:
PATHS OF HOPE: Long Term Survival with Metastatic Breast Cancer
FAQ's on Avastin Withdrawal
The Vitamin D Debate
PARP Inhibitor: Major Triple Negative Breast Cancer Breakthrough
Aspirin and Breast Cancer
THE EDGE CAM
EVIDENCE BASED COMPLIMENTARY SUPPLEMENTATION
LIFESTYLE RISK REDUCTION STRATEGIES
LATE BREAKING NEWS FROM THE CUTTING EDGE:
PATHS OF HOPE:
LONG TERM SURVIVAL OF METASTATIC BREAST CANCER
A Remarkable Journey of Survival
My colleague Stefan Gluck at the Sylvester Comprehensive Cancer Center in Miami, Florida has just published an inspiring case of a 93-year-old woman who was diagnosed with stage I endocrine (ER+/PR+) T2N0M0 breast cancer at the age of 45, then diagnosed 22 years later with metastatic breast cancer (MBC) to the lung, followed by both bone and liver metastases. She survived for 18 years with metastatic lung, bone and liver disease, maintaining a good quality of life (QOL) and remaining mobile (with the help of a cane and walker due to stroke), but in general asymptomatic.
This patient was stable on tamoxifen and then anastrozole (Arimidex) for a total of 12 years, with anastrozole (Arimidex) later changed to exemestane (Aromasin) for another 2 years. Bone metastasis developed 36 years after initial diagnosis, for which she was treated with fulvestrant (Faslodex) and zoledronic acid (Zometa), and then switched to letrozole (Femara). Then liver metastasis was diagnosed 38 years post-Dx, for which she was treated with full-dose capecitabine (Xeloda) monotherapy, discontinued at patient request due to grade 2 HFS (hand-foot syndrome), dying 2 years later.
What is most remarkable is that with multiple visceral metastases (pulmonary, bone, and liver metastases), this patient lived for 18 years with metastatic disease to three sites (two visceral (lung and liver), and in bone) with acceptable QoL, and maintained largely on endocrine only therapy, with chemotherapy only introduced briefly towards end of her treatment.
More Tales of Long-Term Survival with Metastatic Disease
Increasingly,, such cases are not unique: Naruto Taira and colleagues at the National Shikoku Cancer Center in Japan presented a case study of a patient with metastatic breast cancer with pulmonary metastasis who maintained complete remission (NED) for ~10 years (on S-1, a prodrug of 5-FU), while Huang and colleagues at McGill University (Canada) presented a case of a 41-year-old BRCA2 breast cancer patient with lymph node recurrence and bone metastasis who achieved complete remission (NED) for 11 years (Tx with high-dose anthracycline-based chemotherapy followed by autologous bone marrow transplantation).
And there are several case reports of long-term survival with liver mets: Renata Zaucha at the Medical University of Gdansk reported the case of a 66-year-old hormone positive breast cancer patient with multiple liver metastases who survived and remained active at least 11 years (as of last report) free of all breast and visceral disease (after progressing on FEC, she was switched to MPA (medroxyprogesterone acetate) monotherapy which eradicated all cancer disease; while in addition Yoshio Naomoto and colleagues at Okayama University Medical School in Japan presented the case of a 39-year-old woman diagnosed with multiple liver metastasis plus liver dysfunction from breast cancer who survived 8 years with complete remission (treated with chemoendocrine therapy).
A New Vista
These and many other known cases demonstrate that long-term remission, if not (technically) cure, is achievable in the context of metastatic breast cancer, and that in addition, long-term survival with stable metastatic disease is also achievable, allowing for durable and extended prolongation of life with acceptable QoL, even with multiple metastases to lung, liver and bone.
I will finish with repeating the remarks on this theme I made back in 2009:
Lessons from the Past
Increasingly, expert attitudes have changed concerning the curability of metastatic disease and fewer clinical leaders accept the naïve view that patients with metastatic disease invariably die from it, but rather recognize the divergence of this unnuanced view from clinical reality in the fields, therefore marching along with the great Neil Rosen at Memorial Sloan-Kettering Cancer Center who recently at the Controversies in Breast Cancer conference I attended observed radically – but no longer a lonely voice - that he believes in curing metastatic disease, as I do myself. For we now know, from the long-term follow-up series at MD Anderson along with other accumulated data that:
- many patients will die with their disease, but not from their disease;
- some will not even die with their disease (see below);
- still others are currently long-term survivors whose ultimate fate is unknown, but therefore not assured to be from breast cancer mortality;
- still others who are predominantly stable, or at least what I call controlled-progressive - with some degree of progression (sometimes with other tumors in remediation) that can be significantly regressed upon each progression - are, on the odds, highly likely to benefit from what I call the time value of survival: namely, that outcome odds can be improved with each year of additional survival by virtue of the advance of research and clinical developments.
So in this vein, Kathy Miller at Indiana University recently commented on these findings, observing that there is always known to be a "group of patients [who] are potentially cured of their metastatic disease". And the great Gabriel Hortobagyi at MD Anderson, one of the principal collectors of the long-term survival repository, has stated as long ago as 2001: "I also believe that you can cure some patients with metastatic breast cancer" and, based on that data, that for some metastatic disease patients "we can achieve a complete remission". He expands on this widely misunderstood phenomenon: "My group has published about patients who have achieved a complete remission with chemotherapy and are in a progression-free complete remission 20 years later — biopsy proven" and that "for instance, I [Dr. Hortobagyi] have a patient who had a lung metastasis resected in 1968, and she never received any additional therapy. She never had a second metastasis — metastatic breast cancer by definition, pathology proven, and she’s alive and well — 32 years of disease-free survival."
This is not news to clinicians and researchers who deal extensively with advanced disease, as I do. I know, across numerous countries, many 15+ year and a few 25+ year survivors of metastatic disease. As I observed in my Paths of Hope series in these [No Surrender Breast Cancer Support] forums, I even know of long-term survivors of CNS malignancies.
Given this I have proposed every chance I get, with increasing positive feedback and acceptance, to leading clinicians and investigators that we speak of advanced disease, and qualify that when necessary with specification of "with bone involvement" or "with visceral involvement" (lung and/or liver), etc. , or refining as to velocity, or progression or stabilization, or resistance, or refractoriness, etc. in order to avoid the often simplistic, generally unproductive and increasingly counterfactual connotations of somehow "incurable" rather than advanced disease, and consider treating patients with advanced disease with curative intent, not as merely palliative. As one linguist said, language can lift us, and how we speak can drive how we perceive, and also therefore how we act, so let's not let language bully us.
And finally we need to remember that the outliers - the exceptional group of long-term survivors - of today will be the norm of tomorrow.
A Note on Adjuvant CMF for the treatment of Triple Negative Breast Cancer:
A recent (2011, publication ahead of print) study by Shusen Wang and colleagues at State Key Laboratory of Oncology in South China (Classical CMF regimen as adjuvant chemotherapy for triple-negative breast cancer may be more effective compared with anthracycline or taxane-based regimens) retrospectively analyzed 687 TNBC patients undergoing adjuvant chemotherapy (either anthracycline and/or taxane, or CMF), with a median age of 46 years (the range being from 16 to 76 years). The study results found that at a median follow-up of 56 months adjuvant CMF chemotherapy significantly decreased recurrence compared to both anthracycline-based and taxane-based regimens, with a reduction in recurrence of 66% for the CMF treated TNBC patients compared to anthracycline and/or taxane treated TNBC patients, suggesting that CMF regimens for TNBC patients may be more effective in the adjuvant setting than anthracycline- or taxane-based regimens.
A Brief FAQ on FDA's Decision Today to Withdraw Avastin Approval in Breast Cancer
What Was the Decision?
One day in advance of the stated deadline (12/17/10), the FDA has announced that it will revoke the current application for use of bevacizumab (Avastin) in metastatic breast cancer; this application was for first-line use in MBC when Avastin was combined with paclitaxel (Taxol) as per the Kathy Miller E2100 trial.
Is This Decision Final?
No. The FDA decision to withdraw Avastin approval is pending both possible legislative and regulatory-appeal action, and today's decision does not yet in fact withdraw the current indication of use (see below re appeal process). On the regulatory appeal front, I note:
(1) that the manufacturer, Genentech/Roche, has an active / open application pending with the FDA for approval of Avastin not as first-line, but as a second-line therapy for advanced breast cancer after failure of first-line therapy, and the stipulated FDA deadline for that decision in second-line MBC is May 2011.
(2) In Europe (but not in the UK), the EMA is continuing its approval of Avastin in the E2100 regimen of Avastin + paclitaxel (and here in the US the NCCN has strongly supported its continued use); the EMA has judged (correctly) that for Avastin use under the E2100 protocol the benefits continue to outweigh the risks. Furthermore, the FDA has declined to comment or account for how the strict evidence-based European EMA rendered a positive judgment opposite to the FDA's (and sister agency NICE in the UK) negative one despite assessing precisely the same studies and the same trial data, nor how the authoritative evidence-based NCCN organization here in the US has similarly determined the Avastin provides positive survival benefit, and with risks outweighing harms. And just today [12/17/10] the UK CHMP (Committee for Medicinal Products for Human Use) confirmed that the benefits of the E2100 regimen, Avastin in combination with paclitaxel, outweigh its risks and that this combination remains a valuable treatment option for patients with metastatic breast cancer (MBC). It therefore continues to appear that the FDA negative decision on Avastin is in conflict with major non-U.S. evidence-based regulatory authorities, all evaluating the identical set of trials.
(3) Note also that the FDA today (12/16/10) issue a NOOH or "Notice of Opportunity for a Hearing", which provides Genentech with an opportunity for a hearing on a proposal to withdraw Avastin's indication, and Genentech has just this morning confirmed that it will request just such a hearing under NOOH.
(4) The FDA has explicitly confirmed the possibility that if and when Avastin is finally revoked, pending the conclusion of the NOOH appeal process, FDA approval could still be restored for certain subgroups of patients who may benefit (and various experts like Judie Gralow at SCCA have vocally argued that the FDA judgment is in error in failing to distinguish from its overbroad claim that Avastin does not benefit overall survival, certain subgroups for which such benefit may accrue), and in this connection Janet Woodcock, director of the FDA's CDER (Center for Drug Evaluation and Research) program, announced that the " FDA is ready to work with Genentech on any proposals to conduct additional studies in metastatic breast cancer designed to identify responsive tumors ".
What Happens While the NOOH Appeal Process is Underway?
Access to Avastin is automatically extended while these proceedings continue, and I fact with no set date of final decision of the NOOH appeal, so at this time until the NOOH process is completed, Avastin remains FDA-approved for use in combination with paclitaxel for the first-line treatment of any metastatic HER2-negative breast cancer, and in addition oncologists can continue to prescribe Avastin to patients "off-label", and as the FDA's Janet Woodcock has clarified in interview, "This [today's non-approval decision] is a first step in a process and will not have an immediate impact on use of Avastin or the drug's availability."
(Note however, that insurance companies seeking to avoid the associated high costs have already begun restrictions, with Regence Blue Shield (operating in the Pacific Northwest) and certain "blues" (Blue Cross/Blue Shield of Illinois, Mexico, and Oklahoma) publishing new policies restricting or disallowing covered use of the drug).
Was the FDA Decision to Withdraw Approval Justified?
No. This is complex, and I have discussed it elsewhere, and experts have also argued against the legitimacy of the present FDA decision, but in short to my mind:
(1) the approval withdrawal was based on a narrow interpretation of positive outcome as solely overall survival (OS), rather than that increasingly in place in practice currently in the filed, of PFS (progression free survival) as a proximate surrogate of OS (and under this cramp reading many oncology drugs currently in use for metastatic breast cancer would be rendered problematic), and indeed the European EMA as well as the UK CHMP (Committee for Medicinal Products for Human Use) have made precisely the opposite determination from the FDA on this issue, citing that Avastin unambiguously "prolong[s] progression-free survival of breast cancer patients without negative effect on the overall survival", thus accepting PFS as a proximate survival surrogate and acknowledging the importance to the patient's quality of life of an at least additional 5.5 months of survival without recurrence or disease progression.
(2) the FDA failed to weigh in Avastin adjunct (non-primary) use, namely of synergizing chemotherapies given concurrently,
(3) the FDA failed to weigh in the potential of Avastin to interrupt the process of the malignant progression of micrometastases into true (macro) tumors,
(4) the FDA failed to weigh subgroup analysis of defined subpopulations that may receive positive outcome benefit, and
(5) the FDA failed to weigh in the compassionate use of Avastin in populations that have proved refractory to previous traditional lines of treatment OR that have a narrow spectrum of therapeutic options available to them (as in TNBC).
The Vitamin D Debate
The Institute of Medicine (IOM) Dietary Reference Intakes (DRIs) for Calcium and Vitamin D: A Brief Critical Appraisal
Medical advisor, No Surrender Breast Cancer Foundation (NSBCF)
in behalf of NSBCF, dated: 12/01/10
On November 30, 2010 the Institute of Medicine (IOM), of the National Academy of Sciences (NAS), issued a report on Dietary Reference Intakes (DRIs) for Calcium and Vitamin D hereafter, IOM DRI-CVD Report):
1. claiming that most people in the U.S. and Canada should be able to obtain sufficient amounts from their diet to meet Dietary Recommended Intakes (DRIs), assuming no other vitamin D was synthesized from the sun in the skin;
2. raising the DRIs of vitamin D to 600 International Units (IU) for ages 1-70 and to 800 IU for >= 70, compared to the previous DRIs of 200 IU for ages 1-50, 400 IU for ages 51-70, and 600 IU for ages 70 and older;
3. raising the Tolerable Upper Limit (TUL) from 2,000 IU to 4,000 IU for adults.
But our critical appraisal, conducted as a special service for NSBCF, of the full report (999 pages) exposes a number of critical problems and limitations and it is further disturbing IOM did not see fit to include on the investigating and authoring panel any Vitamin D expert, and although some experts provided review commentary, they did not see or review the final report, and the commentaries of these reviewers appear to have been suppress by IOM (resulting in the filing of a federal Freedom of Information (FOI) request for the release of the 14 reports from Vitamin D experts, including from Robert Heaney at Creighton and nutrition expert Walter Willett at Harvard, among several others and many of whom have expressed strong disagreement with the IOM findings and recommendations).
Minimal Optimal Levels: IOMs Claims re Insufficiency and Deficiency are Insufficient and Deficient
First, re claim #1, The IOM statistics are against the weight of the evidence to date, which has established across numerous studies that:
* At least 50% all children in the US and Europe are Vitamin D deficient (deficiency defined as <20 ng/ml)) at some time (three out of every four people in the US are deficient), with virtually everyone being vitamin D insufficiency (defined as 20–31 ng/ml), and with Levels of Vitamin D declining by 22% in the US over the last 10 years, and culminating in the fact that as of last year, 70% of whites, 90% percent of Hispanics, and 97% of African-Americans in the US had Vitamin D insufficiency [per John Adams and Martin Hewison at UCLA, among others].
* Worldwide, between 30 to 80% children and adults are deficient or insufficient even for those residing near the equator .
These and other studies suggest that the minimal acceptable level is 50 ng/ml (125 nmol/L), and Robert Heany and colleagues in a recent study reviewed five studies in which both cholecalciferol (aka, Vitamin D3) and 25(OH)D levels were measured, finding that the body only reliably begins storing cholecalciferol/ D3 in fat and muscle tissue if until 25(OH)D levels rise > 50 ng/ml (125 nmol/L), with chronic substrate starvation of Vitamin D3 at levels below 50 ng/ml, suggesting that 25(OH)D levels should be between 50–80 ng/ml (125–200 nmol/L regularly, across all seasons and verified by multiple sequential readings (pregnant and lactating women need at least 5,000 IU/day, not the IOM prescribed amount of 600 IU, as recently suggested in the studies from Bruce Hollis at he Medical University of South Carolina). And Laura Hall and colleagues at UC-Davis found that the amount of vitamin D intake needed to maintain target serum 25(OH)D concentrations in subjects with low sun exposure and dark skin pigmentation is substantially higher than current recommendations (1300 - 3100 IU//d just to rise above deficiency and insufficiency), confirming the need for adequate supplementation as also found in the just published Ontario Women's Diet and Health Study of 4000+ women using the Ontario Cancer Registry that found, not for the first time, that Vitamin D supplementation was independently associated with reduced breast cancer risk. This is further supported by the recent meta-analysis by the Chinese Academy of Sciences of 37 studies of vitamin D and calcium in the prevention of breast cancer, finding strong evidence that vitamin D and calcium have a chemopreventive effect against breast cancer. And note in addition that benefits of course span breast cancer, as seen in the recent results from Sharif Mohr and colleagues at UCSD who found in their ecological study that countries with low solar UVB irradiance and/or low estimated mean serum 25(OH)D levels generally had higher age-standardized incidence rates of brain cancer.
Vitamin D3 Toxicity: IOM Tolerable Upper Limit - No Basis in Fact
The IOM states that vitamin D toxicity might occur at an intake of 10,000 IU/daily (250 micrograms/d) - and note this is dispute on critical review of the aggregated data to date - although no compelling evidence is cited, so that the posited Tolerable Upper Limit (TUL) of 4000 IU as prescribed by the report appears wholly arbitrary. This is verified in interview with Catherine Ross, Chair of IOM now at Pennsylvania State University and one of the report authors, when asked that if IOM believes as they state that toxicity might set in at 10,000 IU or above, why did they set the Tolerable Upper Limit (TUL) to 4000 IU, she replied: " We needed to take a cautious approach because we're looking to reduce risk to the public. There's evidence of harm at the level of 10,000 IU. So we used 4000 IU per day. " But this is of course is manifestly capricious and not evidence-based, and directly contracted by robust studies: dose ranging studies in healthy adults revealed that 10,000 IU of vitamin D3 daily for five months did not cause any toxicity on calcium metabolism, and an independent study found that even up to 40,000 IU of vitamin D daily for one month had no adverse effect on calcium metabolism [as per Reinhold Vieth and others], suggesting that vitamin D3 is very safe, with a wide therapeutic index, and further suggesting that conservatively it is reasonable to increase the tolerable upper limit (TUL) for vitamin D to at least 10,000 IU/day.
Vitamin D and Breast Cancer: IOM's Narrow Scope or Tunnel Vision
The IOM in their report focused, as they did in their previous report of 1986, on bone health ignored thousands of studies on the benefits of optimal Vitamin D3 levels on cardiovascular, metabolic, oncological, neurological, infectious, and other dimensions of human health. The aggregated data from multiple studies has shown that to optimize beyond just bone health, supplementing is required, given that it is virtually impossible to significantly raise vitamin D levels when supplementing at only 600 IU/day (15 micrograms), IOM prescribed amount, and it has been found that for instance pregnant women taking 400 IU/day have the same blood levels as pregnant women not taking vitamin D, suggesting that 400 IU is a subtherapeutically small dose in pregnant women, and the findings of Bruce Hollis suggest that has shown that 2,000 IU/day does not raise vitamin D to adequate levels in either pregnant or lactating women. And I have already discussed elsewhere the evidence of benefit of optimal Vitamin D3 supplementation based on a target of at least 66 ng/ml for breast cancer populations both for favorably affecting recurrence risk and bone health (as suggested by the Pamela Goodwin Canadian study reported at ASCO 2008 that survival outcomes as measured by distant disease-free survival (DDFS) and overall survival (OS) were significantly worse in women with deficient Vitamin D3 levels, independent of age, BMI, insulin, T and N stage, ER status and grade, and, importantly, these adverse survival outcomes were not significantly modified by adjuvant chemotherapy nor by tamoxifen, confirmed in their follow-up JCO 2009 report, where low plasma levels of vitamin D were again significantly associated with an increased risk of distant recurrence and death regardless of individual tumor-related and treatment-related factors, with the observations that (1) deficient levels of vitamin D were associated with higher-grade more aggressive tumors, and that (2) Vitamin D deficiency was associated with insulin resistance, a known adverse prognostic factor in breast cancer. And after their longitudinal evaluation of vitamin D plasma levels during anthracycline- and docetaxel-based adjuvant chemotherapy in early breast cancer patients, Santini and colleagues in Italy concluded that oncologists should assess circulating levels of vitamin D as 25(OH)D in all early breast cancer patients before starting adjuvant chemotherapy, and to supplement patients with adequate doses of vitamin D before and during chemotherapy.
And note that the URMC James P. Wilmot Cancer Center retrospective study of 224 women with breast cancer found that 66 percent were either severely deficient or moderately deficient of vitamin D. Similarly, this year Xavier Noques and colleagues at the University of Barcelona found a high prevalence of commonly unrecognized Vitamin D deficiency in women with early breast cancer (EBC) treated with aromatase inhibitor (AI) therapy, confirmed by Nicola Napoli and colleagues at Washington University who found the same high prevalence in postmenopausal women with non-metastatic breast cancer on entry to AI therapy, and associated in particular with musculoskeletal complaints. In this connection, Qamar Kan and colleagues at the University of Kansas investigated the effect of vitamin D supplementation on serum 25-hydroxy vitamin D levels, joint pain, and fatigue in women starting adjuvant AI (letrozole (Femara) therapy for breast cancer, finding that after 16 weeks of AI therapy, more women with 25(OH)D levels > 66 ng/ml (median level) reported no disability from joint pain than did women whose 25(OH)D levels where <66 ng/ml, suggesting that active intervention with optimal Vitamin D3 supplementation can protect against AI-induced disabilities and hence increase compliance with AI treatment.
The IOM DRI-CVD Report is methodologically flawed and against the weight of the cumulative evidence to date, and Vitamin D3 supplementation and optimal target 25(OH)D recommendations, as detailed in the latest version of the Edge-CAM regimen, therefore stand unaffected and women who wish to engage on evidence-based breast cancer preventive interventions and patients with breast cancer may reasonable and safely follow the guidance presented under a highly favorable benefit-to-harm ratio.
Major Breakthrough in the Treatment of Triple Negative Breast Cancer
1. A recent updated analysis of the original BS-201 PARP Inhibitor (PARPi) Trial continues to show an exceptional outcome benefit to the addition of the PARP inhibitor (PARPi) BSI-201 to the gemcitabine-carboplatin chemotherapy backbone, and although these are interim and not final results which could change in either direction, at this time the interim findings support the provisional conclusion that the PARP inhibitor essentially - and amazingly - doubles overall survival (OS) - not just PFS (progression-free survival), also improved, meaning reduction in the risk of recurrence - in the triple negative breast cancer population, and at this time sustains a 50% reduction in the risk of death.
[Disclaimer: we must wait the final results of the trial to see if these dramatic, first-ever results, are sustained at the same or reasonably comparable levels as now seen with the interim findings].
2. These rather stunning survival outcomes were furthermore accompanied by highly impressive response rates: the overall response rate (ORR) was high, at 48% of patients achieving complete response (CR) or partial response (PR) which is three times as high as that obtained with chemotherapy alone, and with another 14% achieving stable disease (SD, for 6 month or greater), yielding a clinical benefit rate (CBR) of 62% (CBR = CR + PR + SD), and with no significant additional adverse effects from the addition of BSI-201 to the chemotherapy backbone (impressive, remembering that non-toxic agents are notoriously hard to come by in oncology).
3. The accrual and progress of the trial has proceeded at extremely rapid pace, well beyond expectations, and the best estimation, based on feedback from investigators, is that this Phase III trial given the pace, will come to completion as early as this (First) Quarter of 2010. Interview statements - but not officially posted NCI protocol data - suggests that there are just 40 patients remaining requiring trial accounting, 20 in the PARP inhibitor arm, and another 20 in the chemotherapy only arm, confirming that completion is very close, and that a First Quarter 2010 estimate is plausible on the progress to date.
4. Finally, on the regulatory front it also appears there is accelerated progress: the PARP inhibitor BSI-201 has been granted on Fast Track Designation by the FDA, very good news for patients: FDA Fast Track means that, against standard requirements, the agency will accept initial late-stage data instead of waiting for entire Phase III clinical trial results, something that is done when (1) a proposed agent is intended for treatment of a serious or life-threatening disease - a status now accepted by the FDA for mTNBC (metastatic TNBC) - and (2) demonstrates the potential to address unmet needs for such a condition. Based on this status and on a review of documents filed in the FDA regulatory pipeline process, it is now estimated that BSI-201 may actually become commercially available - and hence available to all mTNBC patients in clinical practice without being on any clinical trial - at year's end (mid November to mid December, best estimate).
Commentary: Missed, and New, Opportunities
My own sense of the ASCO BSI-201 PARP trial is that on the contrary that it may have underestimated the true benefit; this perspective stems from my own TNBC review and research which on the cumulative evidence suggests that maximal benefit of PARP inhibition is accrued when it is concurrent with a strongly genotoxic (DNA-damaging) regimen, and although carboplatin is genotoxic, as are all platinum agents, gemcitabine (Gemzar) is not, and I believe this represents a lost opportunity. Indeed, I have on several occasions advocated in this context the omission of gemcitabine (Gemzar) altogether in the PARP context, in favor either of (1) a more optimal dose of carboplatin - I consider carboplatin AUC=2 as substantially sub-optimal, and would have deployed at least AUC=6 even up to AUC=7.5, or (2) substituting another genotoxic agent instead of gemcitabine (Gemzar), my choice being an anthracycline (preferably the pegylated liposomal Doxil or Caelyx. Given this limitation - which in all fairness the principal investigator Joyce O'Shaughnessy has acknowledged (due to expediency, not oversight) - of the failure to maximize the potential synergy of genotoxicity and PARP inhibition, in my mind therefore the trial's findings are actually therefore even more impressive by extrapolation, ad I would predict that an all-genotoxic chemotherapy + PARP inhibitor (BSI-201) will achieve significantly greater outcome benefit than even doubling of survival and halving of mortality for metastatic TNBC patients.
And this is where the opportunity can be regained, in the earlier availability of BSI-201 by this year's end, since at that point clinicians are no longer constrained to conform to the trial protocol's chemotherapy regimen, and I would advocated for them to instead adopt a more optimal all-genotoxic regimen which should translate to even more dramatic outcome gains. I am also of the opinion for molecular and other reasons that the failure of response of the PARP inhibitor regimen for some patients may be due in part to the absence of pure genotoxicity in the combination regimen, which therefore I am hoping innovative oncologists will recognize and overcome through appropriate DNA-damaging agent substitutions for the non-genotoxic gemcitabine (Gemzar) component, allowing more patients to be responsive to this breakthrough for metastatic TNBC disease.
Come The Revolution
Finally, remember the last time we heard of a 50% gain in breast cancer survival? That was with the revolutionary, practice-changing (virtually overnight) findings from Dennis Slamon's trial on trastuzumab (Herceptin) for HER2-positive disease, heralded correctly as the greatest breakthrough in breast cancer treatment since tamoxifen 30+ years ago. I believe an optimally genotoxic PARP inhibitor therapy can do for triple negative disease what Herceptin did for HER2+ patients, and the provisional results of the Phase III PARP Inhibitor to date appear to support and imply that contention.
These are the kind of times that researchers like me - and dedicated and brilliant investigators, and soon also patients - live to see as their reward for their commitment.
Dissecting the Holmes Study: The Good, Bad, and Ugly
The Michele Holmes study was a prospective observational study based on responses from about 4000+ registered nurses in the Nurses’ Health Study (NHS), measuring breast cancer mortality risk according to aspirin use frequency (number of days per week) at least 12 months post initial diagnosis. The study found that aspirin use after a breast cancer diagnosis was associated with a decreased risk of (1) distant recurrence, (2) mortality from breast cancer, and (3) mortality from any cause.
Muddles and Limitations, and Some Praise
As to staging, all subjects were non-ABC (non-Advanced Breast Cancer) stage diagnosed at stages I, II, or III, for usage greater than 2 days weekly, the association not differing appreciably by stage, menopausal status, body mass index, or estrogen receptor status (both ER-positive and ER-negative subjects benefited equally. Given this later observed fact, namely the ER status-independence of aspirin benefit, the authors display their first signal of well-intentioned but misguided fuzzy thinking, as they argue that "aspirin may improve survival through various mechanisms . . . may lower serum estradiol"; this is inconsistent on the face with their own datum of ER-independence.
Now our always shrewd Snhb raises some reasonable points of skepticism, and the study does suffer from the typical limitations of all observational studies, including recall, since all data on aspirin intake, treatment, and distant recurrence was self-reported. In addition, as she also points outs, there is no data on aspirin doses used, and the authors acknowledge and partially address these limitations, not wholly successfully. In a rare candidness, they admit that if there was a dose response effect, the effect size, in their study they speculate, might be diminished because frequent aspirin users are likely to be low-dose users for cardiovascular prophylaxis. True enough, but backward: that honest admission in fact is wrong-headed and oddly misreasoned, since the significant relative risk reductions obtained would be more striking, not less, if there was indeed a large pool of low-dose users. So in a sense, despite study limitations, their study is in some ways somewhat better than it first appears, and somewhat better than even the authors acknowledge.
To my mind, also to their credit - and answering snhb's sharp question re other potentially confounding factors - they correctly - and adequately - adjusted for an uncommonly large battery of these covariates as they are called, including breast cancer stage, diet, physical activity, body mass index (BMI), weight change, reproductive factors ( age at first birth and parity, oral contraceptive use, menopausal status and use of hormone replacement), smoking, protein and energy intake, and even calendar year (to adjust for secular trends), and critically, too, for all treatment (chemotherapy, endocrine therapy, and radiotherapy) received [from raw data I extracted from the trial cohort].
So there are indeed some study limitations and some partially countervailing strengths, along with several instances of seemingly wooly-headed reasoning, leaving the study highly suggestive of a trend toward significant benefit from at least modest aspirin usage, but certainly not definitive, not of course to have been expected from an observational study which definitionally cannot render a dispositive finding of fact.
There are to my mind two other questions the authors should have raised but didn't, the most obvious being the inconsistency with another study of this issue using the same NHS II trial data, that of Heather Eliassen's prospective analysis of 112,000+ premenopausal nurses, reported last year, following similar methodology but failing to find any strong associations between aspirin, NSAIDs, or acetaminophen and breast cancer risk in the quite large prospective cohort of premenopausal women studied, after 14 years of exposure information and follow-up; there the authors concluded that "chemopreventive use of aspirin or other NSAIDs for breast cancer among premenopausal women is not warranted". This 2009 study, like the Holmes 2010 study under discussion, argue somewhat beyond the facts albeit in different directions, and both studies exhibit an unflattering "tunnel" effect - no single observational study can be dispositive, but that is "noise", an irrelevancy unless we forget all precepts of evidence-based methodology; we are after what the balance of the evidence warrants, across multiple observational cohorts, and multiple robust systematic reviews and meta-analysis. The second and related question side-stepped, by this and most other studies claiming to adjust for treatment effect, is that not all treatments are created equal: we may have a simple ACx4 regimen, or an ACx6, or AC ---> T (paclitaxel typically), or tamoxifen, or an AI, or trastuzumab (Herceptin), quite a range yet leaving opaque the confounding effects of diverse, and diversely scheduled (standard vs. dose-dense for instance), regimens. And in addition we do not know whether aspirin may benefit more with say genotoxic chemotherapy versus standard cytotoxic therapy, or anti-angiogenic therapy, and so on. But all studies share this limitation and oversight, so the Holmes studied should not be unduly disadvantaged vis a vis others, and given that aspirin still seems to have a sufficiently consistent effect across regimens (as I suggest below in my brief summation of systematic reviews and meta-analyses to date), this critique is not fatal to the study's intent and outcome reported.
But What Really Counts . . .
What counts ultimately is not "the evidence" from one or another study, but the "weight of the evidence" as accumulated over robust and methodologically properly cross-analyzed and pooled studies, individually subjected to critical appraisal. Looking at "clinical query" data culled from systematic reviews and meta-analyses, aggregating the best of these shows a clear trend in support of at least a modest positive and significant chemopreventive benefit from low to moderate NSAID use including aspirin (the NHS data can obviously be read any way, since two studies I discussed above reached conflicting conclusions, and in any case in my view nurses are a somewhat "unnatural" cohort not truly representative of the general random population, being more medically savvy about prophylaxis, and more adept in navigating their way to higher quality care), with the cumulative cross-data suggesting between approximately one tenth to one third risk reduction of breast cancer incidence.
What the Weight of the Cumulative Evidence Says
Thus, assessment of 34 case-control and 13 cohort studies showed that the relative risk of breast cancer was 0.77 in those taking aspirin (meta-analysis of González-Pérez et al.), and in meta-analysis of eight case-control and six cohort studies (Khuder et al.) the relative risk of breast cancer was 0.82 while data across 20 observational studies suggest that NSAIDs both offer some protection against breast cancer and might be of benefit to women with active malignancies (meta-analysis of Agrawal et al.). And across all these dozens of trials, we have methodological re-assurances of robustness of observed findings, including (1) the cancer reduction seen is comparable for both men and women, (2) elective termination of prophylaxis is seen to be followed by a rapid loss of protection, (3) several NSAIDs present with effects similar to aspirin, while non-NSAID analgesics not affecting COX pathways consistently fail to reduce cancer incidence.
And although from the Holmes study it may still be premature to conclude more narrowly about benefits of mortality reduction and reduction in distant metastases, as opposed to chemoprevention of breast cancer disease, nonetheless - without being dispositive - we have accumulating converging data of the plausibility of just such positive outcome, and there is mounting and highly compelling data on the intimate and highly adverse effect of inflammation in carcinogenesis, tumorigenesis, malignant progression and the metastatic cascade, with some frontier edge evidence of the tight interweaving of microRNA pathways (which I will soon be discussing in connection with epigenetic reprogramming), free radical pathways, and both cytokine and p53 pathways on the one hand and inflammation in cancer pathogenesis on the other, operating at highly primitive underlying strata. Like angiogenesis, inflammation is rapidly being shown a core causal pathway of cancer, and aspirin in particular not only effects inhibition of COX-2, but also inhibits the activation of NF-kB, one of the most fundamental pathways of malignant progression, as well as effecting the upregulation of powerful tumor suppressor genes, especially (T)p53 and BAX, while downregulating antiapoptotic genes such as BCL2, and aspirin also effects DNA mismatch repair processes, all this cumulatively demonstrating antitumor mechanisms that are COX-2-independent.
On the Dark Side, and Back
Aspirin, even at the low doses appropriate for vascular use, roughly doubles the incidence of gastric bleeding, but as I have argued elsewhere on these forums, this is a manageable adverse event with proper caution. And as evidence was accumulated to show that aspirin and NSAID-related risk of GI bleeding is increased in the presence of H pylori infection, just underlines the case I have already made for licorice-DGL type therapy which is eradicative against this microbe. Nor, as I have also repeatedly argued, should aspirin benefits be seen as contraindicated in the presence of anti-platelet therapy (like Plavix) or anticoagulant therapy (like warfarin (Coumadin)), since that is what PT/INR testing is for after all, and dose adjustment of either class of active agents should be made as dictated by the INR level with concurrent aspirin, assuring a target of 2.0 - 3.0.
In addition, although evidence for a relation between cerebral bleeding, and therefore the risk of hemorrhagic stroke, and aspirin is less certain, the seminal meta-analysis by Jiang He and colleagues of data from 13 RCTs (mean age between 60 and 70) found the relative risk of a cerebral bleed from aspirin to be between about 1.4 and 2.2, with an absolute incidence of bleeding attributable to the drug of 1.2 per 1000 persons per year, non-trivial but still relatively small, and note that cerebral bleeds do not inevitably lead to hemorrhagic stroke, and the most reasonable estimate of the risk of hemorrhagic stroke associated with the use of aspirin in primary and secondary prevention patients (from Gorelick and Weisman) is just 0.2 events per 1000 patient-years, which in fact fails to achieve statistical significance, and as the large Rotterdam Scan Study reporting last year concluded on this risk, the beneficial effects of aspirin and related agents against myocardial infarction or ischemic cerebrovascular disease have been shown to outweigh any risks of bleeding, at whatever locus or origin, so that the relatively small adverse events of aspirin use even in their totality must be balanced against its multiple possible benefits in particular for cancer prophylaxis of multiple different malignancies types, and cardiovascular disease and stroke.
A Question of Balance
Aspirin therefore appears to have on robust evidence relatively benign adverse effects, and may also help chemopreventatively reduce the risk of colon and breast cancer on the balance of the best evidence to date, along with significant reduction of cardiovascular disease and stroke, and although benefits of aspirin to other cancer sites are less consistent than in the very strong colorectal studies, nonetheless aspirin use has also been associated with reduced incidence of multiple other malignancies, including those of the esophagus, stomach, lung, bladder, ovary, prostate, mouth, and skin.
Into the Mire: The Vexing Question of Dosing
The studies have been inconsistent and not dispositive on what the optimal dosing level should be for use of aspirin in the cancer setting, with some studies suggesting low dose (80 - 81mg/d), others moderate dose (>100mg/d and less than full aspirin (300 - 325mg/d, and still others for specialized cases - adenomas - suggesting that >300mg/d is required. How to decide? For me, I am reasonably persuaded by the compromise demonstrated and suggested by Lisa Swaim and Robert Hillman who found that there was significantly less inhibition of platelet function with what is known as the WHS-dose (from the Women's Health Study), which is 100mg evey other day, rather than the customary U.S. dose of 81mg/daily ("baby" aspirin). Finally, lower doses may be highly efficacious in the context of several components of the Edge-CAM regimen which themselves are COX-2 inhibitory, among dozens of other positive benefits, but nonetheless it strikes me favorably the wisdom of an alternative day low-dosing (81 to 100mg every other day; full-dose aspirin on alternative days can be used if one is not on high-dose curcumin (> 1500mg daily)) to minimize adverse impact to platelet function, at least until further trials provide more dose-specific definitive evidence.
So this one study (Holmes et al.) has not settled the question, but the evidence in toto - from the best of clinical systematic reviews and meta-analyses - has spoken, and unambiguously. Many of us will long regret not leveraging the importance of anti-inflammation therapy against a disease, cancer, that increasingly is shown to be intimately and inexorably tied to the adverse inflammatory process, as we must surely regret not heeding the wisdom of HD-D3 for example.
Time for preemptive action.