Numbers Matter: Society and CancerLinQ

Numbers Matter: Society and CancerLinQ

Clifford A. Hudis, MD, FASCO, FACP

@CliffordHudis
Oct 04, 2013
This week I had the good fortune to attend the 10th Main Meeting of the Early Breast Cancer Trialists’ Collaborative Group in Oxford, England. Known widely as “the meta-analysis” or as the “Oxford Overview,” this effort has relied on very large numbers (of randomly assigned patients) to fuel the transformation of post-operative adjuvant systemic therapy for breast cancer around the world.

 

In a nutshell, the earliest randomized systemic trials in breast cancer were too small and underpowered to overcome the random fluctuations in outcome. As a consequence, for years one could cite a paper showing that adjuvant chemotherapy using, for example, cyclophosphamide, methotrexate, and 5-fluorouracil (CMF), was lifesaving; or cite a similar paper suggesting that this regimen actually raised the risk of disease recurrence.

The Overview—by incorporating almost every randomized trial of a specific adjuvant therapy (CMF, tamoxifen, and so on)—generated huge numbers that enabled far more precise conclusions. From that, we learned to give chemotherapy broadly, and use tamoxifen when hormone receptor assays were not negative. (A double negative that allows for treatment when the receptor status is unknown, as still can happen in some circumstances.)

Sometimes criticized as generating a “one-size-fits-all” approach to breast cancer, the Overview has been a force for quality and standard practices. I may have romanticized the meeting and the work because of its focus on informing practical conclusions and pragmatic guidance for clinicians, but so be it. We are in this to improve and save lives, and the Overview was one of those projects with tangible results and human impact.

For years, the Overview updated every five years with appreciable practical implications for clinical practice. I remember when the 1995 Overview first demonstrated that we should give tamoxifen for pre-menopausal patients with hormone responsive disease, for example, and we immediately changed practice in North America, undoubtedly saving lives.

The underpinning of the Overview is the availability of large (and small), properly conducted, prospective, randomized clinical trials with survival endpoints. While these studies have been conducted globally, one key source of support for many has traditionally been the American government, through its funding of the NIH and the NCI. As I write this, after years of slow and steady erosion of their financial situation, they are suddenly at a full stop because of (from the outsider’s perspective) the astonishingly short-sighted decisions being made in Congress. The hidden long-term costs of this week’s government shut down are truly incalculable. There are no numbers to completely and accurately quantify the lost productivity, the projects not started, the experiments never done, the amendments not pursued, the patients not enrolled, and the advances delayed or never realized.

One specific consequence of decreasing support for clinical trials will be our inability to generate the primary data that enables life-saving efforts like the Overview. Many will argue that the era of big, unselected studies that comprise the Overview has passed. In some situations they are right, and the number of clinical trials testing post-operative adjuvant therapy in early-stage breast cancer has dwindled. But, even in, for example, the HER2 positive subpopulation, the Overview enables cross-study comparisons and collaboration focused on biomarkers and toxicities.

Imagine a future where the studies are never done? Then try to imagine how we make the next generation of clinically meaningful and life-saving progress. Our current impasse does not make anything better or easier, and as much as it threatens health now, it does so even more in the future in terms of lost, missed, and delayed opportunities.

This brings me to one of the promises of CancerLinQ. As you will soon formally hear, your Board of Directors has taken the bold step of investing ASCO’s time, talent, resources, and money in the building of version “1.0” of CancerLinQ. The first version will offer basic functionality, but using a modular approach we will add more components and disease areas in the future. Behind the scenes, your Society is generating new concepts, imagining new functions, establishing partnerships and collaborations, and trying to both envision and build a program that is first and foremost designed to help us all deliver the highest possible quality of care to the broad range of patients ASCO members touch.

CancerLinQ offers a way forward that may overcome some of the challenging numbers we now face. Only about 3% of adults with cancer enroll in clinical trials. How do we learn something from the 97% receiving routine care? The clinical trials enterprise is expensive (perhaps more than it needs to be), slow (largely because of important but resource-consuming regulations, oversight, review, and reporting), and sometimes duplicative. But like the old joke about the food at a Catskills Resort (“The food is so bad,” says one patron. “Yeah, and the portions are so small,” says the next.), one thing worse than the current system is another that generates even less data from fewer studies. Relying on CancerLinQ to deliver post-marketing surveillance and post-study “real-world” trend confirmation, perhaps we could conduct fewer studies, offer more accelerated approval, and then truly see what happens in the clinic in broad patient populations not necessarily typical of trial participants.

The scenario in which CancerLinQ directly supports conventional prospective clinical trials is obtainable and within reach. Given the painful numbers we are seeing daily, we should collectively think about how to compensate in new and creative ways. If you see a potential use for CancerLinQ in maintaining and increasing the delivery of high-quality care while accelerating the pace of discovery, please let me or any of your Society’s leaders know. Perhaps each of us is just one number (a “1”), but together we are many multiplied by the power and impact of our collaborations and creativity.

The same way that the Overview pulled hidden truths out of the small numbers of patients in early randomized trials and changed the course of breast cancer treatment, I hope that CancerLinQ will identify unseen truths from the massive numbers generated by all of the care we deliver and record daily. That will be an important advance and contribution.

Disclaimer: 

The ideas and opinions expressed on the ASCO Connection Blogs do not necessarily reflect those of ASCO. None of the information posted on ASCOconnection.org is intended as medical, legal, or business advice, or advice about reimbursement for health care services. The mention of any product, service, company, therapy or physician practice on ASCOconnection.org does not constitute an endorsement of any kind by ASCO. ASCO assumes no responsibility for any injury or damage to persons or property arising out of or related to any use of the material contained in, posted on, or linked to this site, or any errors or omissions.

Comments

Bhawna Sirohi, MBBS, FRCP

Oct, 12 2013 6:31 AM

Totally agree with Dr Hudis that the indivdual patients data meta-analysis generated by the Oxford group today is the gold standard for us clinicians to be able to implement in clinical practice for the good of our breast cancer patients. 

We need a more coordinated effort in other disease types to make this happen both for early and advanced cancers and ASCO could spearhead this with the help of oncologists. 

 


Advertisement
Back to Top