Cancer Staging: Important Work of the AJCC

Jeffrey E. Gershenwald, MD, FACS, Dr. John M. Skibbter Professor, Dept. of Surgical Oncology; Professor, Department of Cancer Biology; Medical Director, Melanoma and Skin Center, The University of Texas MD Anderson Cancer Center, Houston, TX; Member, American Joint Committee on Cancer (AJCC) Executive Committee; Member, AJCC Eighth Edition Editorial Board (Melanoma Expert Panel & Content Harmonization Core); Email:

Q: Staging of cancer is traditionally important. What is the AJCC, who funds it, what does it do? How essential is it for the proper care of cancer patients?

A: The American Joint Committee on Cancer (AJCC) is a multidisciplinary organization with roots dating back to 1959 with the goal of developing and promoting national standards for cancer staging in the U.S. Currently, 22 member organizations spanning multiple disciplines – cancer care, registry, cancer control, pathology, epidemiology, education, outreach and advocacy – support its mission. The AJCC provides worldwide leadership in the development, promotion and maintenance of evidence-based systems for the classification and management of cancer in collaboration with multidisciplinary organizations dedicated to cancer surveillance and to improving care. Administrated by the American College of Surgeons (ACoS) with governance overseen by representatives from founding and other sponsoring organizations, financial support is also provided by several of its member organizations. Notably, other than its core administrative team, it is a nearly completely volunteer-led organization that receives evidence-based input from hundreds of expert volunteers.

Cancer staging serves as the principal communication tool between physicians and their patients and among physicians for clinical decision making and prognostic assessment. Cancer staging is also used for clinical trial design, eligibility, stratification, and analysis. It serves a critical role as the foundation for reporting on an institutional, regional, state, national, and international registry level to facilitate understanding of the broader cancer landscape. With all of this, cancer staging is a key pillar in translational research.

The AJCC published the First Edition AJCC Cancer Staging Manual in 1977. To maintain relevance, AJCC cancer staging systems have been revised and expanded every 5 to 7 years across myriad disease sites included in the manual. The most recent version – the Eighth Edition AJCC Cancer Staging Manual – was published in late 2016 and implemented nationwide in the U.S. on January 1, 2018. It is likely that other countries will also adopt these new staging systems.

The foundational elements of AJCC staging are disease-site specific and anatomic-based (TNM system). Components of the TNM staging system include T (primary tumor), N (regional node and non-nodal regional disease), and M (distant metastasis). The TNM-based system has been employed globally, and continues to benefit from a 30-year partnership with the Union for International Cancer Control (UICC). Cancer staging is performed for a patient at presentation as well as at varying times during their cancer continuum. Such staging classifications include: clinical and pathological (the two most commonly used), posttherapy/post-neoadjuvant therapy (for patients who receive “upfront” systemic and/or radiation treatment as an initial component of their care), recurrence/retreatment (for formal cancer restaging), and autopsy (aTNM). Patients are grouped into cohorts according to risk into various prognostic stage groups.

Despite its success over the past several decades, TNM/anatomic-based staging systems have been de facto constrained in their ability to accommodate improved understanding of cancer biology. In order to be useful, however, they need to be clinically relevant, reflect contemporary practice, and be optimally refined by iterations as our understanding of a given cancer matures.

In an effort to retain clinical relevance, the AJCC has expanded its principles of cancer staging to include non-anatomic based factors (e.g., Gleason score, PSA, mitotic rate) beginning with the 6th Edition (2002), and in the 8th Edition has integrated molecular signatures into some staging chapters (e.g., breast). Formal AJCC acceptance criteria have also been recently developed to serve as a framework for inclusion of contemporary risk models. These changes are overall reflective of a strategic evolution from population-based staging to a more personalized approach.

Given the rapid advances in our understanding of the clinical, molecular, and immunological underpinnings of cancer across multiple cancer types, it is likely that a less “staccato” approach to cancer staging will be devised and implemented by the AJCC; strategically configured and coordinated iterative or “rolling” updates can more efficiently exploit integration of clinically relevant advances into the cancer staging arena. Failure to maintain relevance in this exciting and unprecedented era of cancer discovery and care will eventually render any staging system obsolete.

Jeffrey Gershenwald’s contact info is included in the author affiliations at the top of this page.

Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Foundation Medicine, CMS, FDA, CAP and Academic Molecular Pathology: A Clash of Values

Karen Kaul MD, PhD, Chair, Department of Pathology and Laboratory Medicine, NorthShore University HealthSystem; , Duckworth Family Chair of Molecular Pathology; Clinical Professor of Pathology, University of Chicago Pritzker School of Medicine; Chicago, IL; Email:

Q: The field of precision oncology continues to seek clear guidance on payment for both molecular testing and resulting therapy. In the US, such guidance often is provided by the federal CMS. As an academic and practicing molecular pathologist, how do you view the recent proposal from CMS?

A: Since the implementation of KRAS testing of colorectal cancers a decade ago, oncologists and pathologists have seen the steady growth and evolution of precision oncology. Assessment of somatic alterations in tumors has become a requirement for appropriate diagnosis for many tumors, and is instrumental in selection of effective therapy for others (1-3). Initially, laboratories utilized single-gene assays, but the explosion in genomic targets needed has led to the implementation of next-generation sequencing to efficiently (in terms of time, cost and the small size of many tissue samples) provide the needed information (4). The laboratory community has responded with published consensus guidelines for validation and performance of NGS on tissue samples, along with published studies demonstrating the high quality of the results generated by laboratories that adhere to these guidelines, operating in a CLIA environment (5-9). These advances have taken place largely through the efforts of practicing molecular pathologists working at academic and large community hospitals across the US and around the world.

Pathology, and genomic pathology, is practiced locally in an individualized manner. Each tissue specimen is examined grossly and microscopically, with appropriate sections selected for immunostains or molecular studies. Samples may be tiny, consisting of core biopsies or fine needle aspirations, so that pathologists must carefully triage the tissue, communicating with oncologists to ensure that key information for that patient can be obtained. Findings are discussed at tumor boards which include an integrated team of professionals. Tumor boards are also valuable opportunities for physicians, and especially residents and fellows, to experience and learn the impact of these new technologies on the patients they care for. As precision oncology continues to advance, we must be sure that the next generation of oncologists, pathologists and other health care professionals are fully fluent in utilization of these approaches.

The recently proposed national coverage determination from CMS is, simply put, bad for patient care (10-11). The commercial vendor who worked with the FDA and CMS to obtain clearance did not develop the methods or quality systems utilized (which are in use in clinical laboratories across the country) but did have the financial resources to accomplish this approval. Few academic and community hospital labs will be able to pursue this approval in this era of cost containment, particularly if reimbursement for these services is further obstructed. Furthermore, the large gene panel that was approved, and proposed to be covered by this NCD, will not be suitable for many samples, many types of cancer, and certainly not biopsies or aspirates. Sending the representative tumor block may negate performance of other studies needed for appropriate patient care. Uncertain reimbursement for local laboratories will preclude performance of these critical studies at many centers, leaving them to await results from a remote laboratory not likely to be able to handle the testing volumes. Advances in the field made by local and academic laboratories will be hindered as institutions seek to trim costs and unreimbursed services. Certainly, the robust academic discussions that are so much a part of modern management of oncology patients, and the training of our residents, fellows, and colleagues, will lessen with the reduction of these important procedures to a send-out commodity. Precision laboratories are fundamental to precision oncology and are an important part of patient care; it is critical that they be performed according to quality standards in a variety of settings and centers, and that they be able to contribute to ongoing advances as well.

There is no evidence that FDA’s newly established Breakthrough Device Program will lead to improved testing quality. In fact, it may be argued that the collective comparisons currently made between labs through the College of American Pathologists and other proficiency testing programs more broadly raise quality, particularly when coupled with adherence to laboratory performance standards for NGS. There is a great need to increase quality improvement opportunities available to labs, and especially to collect outcomes data on the impact of this testing in a way that helps us all improve patient care. The public needs would be served best by insuring reimbursement of NGS to facilitate access to service, and to provide a minimally burdensome program to collect the needed data (about quality and outcome) from laboratories, by building on the current CLIA programs. Data-driven consensus guidelines addressing such issues as appropriate timing, sample types, assay coverage and detection sensitivities should be developed. With only a single reimbursed provider for these services, advances in the field will be stymied.

Karen Kaul’s contact info is included in the author affiliations at the top of this page.


  1. Kaul KL, Sabatini LM, Tsongalis GJ et al. The Case for Laboratory Developed Procedures: Quality and Positive Impact on Patient Care. Acad. Pathol. (2017), 4:1-21
  2. Kaul, K.L. Virchows Arch (2017) 471: 141.
  3. Corless CL. (2016) Next-generation Sequencing in Cancer Diagnostics. J. Molec. Diagn. 18:813-816.
  4. Misura M, Zhang T, Sukhai MA, Thomas M, Garg S, Kamel-Reid S, Stockley TL. (2016) Comparison of Next Generation Sequencing Panels and Platforms for Detection and Verification of somatic Tumor Variants for Clinical Diagnostics. J. Molec. Diagn. 18:842-850.
  5. Gargis AG, Kalman L, Bick DP, da Silva C, Dimmock DP, Funke BH, Gowrisankar S, Hegde MR, Kulkarni S, Mason CE, Nagarajan R, Voelkerding KV, Worthey EA, Aziz N, Barnes J, Bennett SF, Bisht H, Church DM, Dimitrova Z, Gargis SR, Hafez N, Hambuch T, Hyland FCL, Luna RA, MacCannell D. (2015) Good laboratory practice for clinical next-generation sequencing informatics pipelines. Nature Biotechnology 33, 689–693 (2015) doi:10.1038/nbt.3237
  6. Sireci AN, Aggarwal VS, Turk AT, et al. Clinical Genomic Profiling of a Diverse Array of Oncology Specimens at a large academic center: Identification of targetable variants and experience with reimbursement. J. Molec. Diagn. 2017, 19:277-287;
  7. Lih C-J, Harrington RD, Sims DJ, et al. Analytical validation of the next-generation sequencing assay for a nationwide signal-finding clinical trial: molecular analysis for therapy choice clinical trial. J. Molec. Diagn. 2017, 19:313-327;
  8. Deans Z, Watson CM, Charlton R, et al. Practice Guidelines for Targeted Next Generation Sequencing Analysis and Interpretation.
  9. Jennings LJ, Arcila ME, Corless C et al, Guidelines for Validation of Next-Generation Sequencing–Based Oncology Panels; Journal of Molec Diagn, 19: 341-365; DOI:

Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Do you have pain, cancer, or diabetes? Your PBM may now be your doctor for these illnesses.

Charles L. Bennett MD PhD MPP, Smart State and Frank P and Jose M Fletcher Chair, Medication Safety and Efficacy, Smart State Center of Economic Excellence, University of South Carolina and the Hollings Cancer Center at the Medical University of South Carolina; Charleston, South Carolina; Email:

Q: The opioid epidemic is now a public health emergency in the United States. Diabetes is now the leading public health emergency worldwide. We recently (June 21, 2017) discussed how Pharmaceutical Benefit Managers (PBMs) have developed from 1968 to controlling not only pricing, discounts, and drug selection for > 250 million Americans, but are also threatening to become the prescriber. Does this raise heightened concerns as the country faces two public health epidemics of opioids and diabetes, in particular?

A: You and I continue to believe that medical care responsibility logically devolves to the doctor who cares for the patient and is accountable for management, treatment, and outcomes. We are increasingly wrong. Since medicines have grown in cost and pharmaceutical costs continue to increase, profit-oriented businesses muscle the patient and the doctor (and stay tuned, even the hospital) out of the way. Pharmacy Benefit Manager (PBM)’s plans now often replace physician choice, and if the physician’s choice is expensive, they replace it with more profitable alternatives, a practice known as non-medical switching. These concerns are magnified as one of the largest PBMs (CVS) seeks to merge with one of the largest health insurers (AETNA), ultimately increasing PBM involvement where patients receive care. Just as worrisome is the “invisible” hand of PBMs in restricting treatment options for chronic pain often experienced by cancer patients and others.

Can you remind us what are Pharmaceutical Benefit Managers (PBMs) and how did they get here? PBMs were developed with lofty goals to act on behalf of insurance companies and payers to maintain or reduce prescription costs while concurrently improving outcomes. These “middle men” corporations are now the principal players in our health care system, controlling at least 80% of drug benefits for 260 million Americans. The initial goal has gone astray. What started as a value-based strategy, allowing for individual case decision exceptions, morphed into blatant non-transparent and cost-based guidelines selecting from restricted lists of drugs based on the lowest cost and highest profit to the PBM, onerous prior authorization and step therapy processes, convoluted co-pays and tier formulas that disregard patient need and clinical judgment, and (in the future) requirements about where patients will receive their care.

How are PBMs handling the opioid epidemic? The impact of this ill-suited logic for pharmaceuticals is clearly demonstrated by the current opioid epidemic in the United States and the role of PBMs in driving patients to opioids, away from abuse-deterrent form (ADF) and less addictive forms of opiates through formulary and pricing strategies. “The three biggest PBMs cover no more than three FDA-approved abuse-deterrent formula opioids (Oxycontin ER, Embeda, Hysingla ER) and do not cover 7 FDA-approved abuse-deterrent formula opioids (Targinig ER, MorphaBond ER, Xtmapza ER, Troxyca ER, Arymo ER, Vantrela ER, and Roxybond). CVS Caremark, which has nearly 90 million members, doesn’t cover a single one. Only one third of Medicare prescription drug plans will cover a more pricey but less addictive opioid, and require prior authorization for other opioid alternatives. These pharmacies do, however, cover the cheaper generic opioids which are readily diverted to abusei. An Institute for Clinical Economic Review studyii found that the use of the abuse-deterrent formulation version of OcyContin could prevent 4,300 cases of abuse and save $300 million in medical costs over a five-year period. In fact, in response to PBM policies not to cover less addictive opioid alternatives led Dr. Milton Packer, the distinguished cardiologist, to conclude that payers are the leading cause of death in the United Statesiii.

How are PBMs handling oncology patient concerns? The exclusion of the patient and physician in drug choice and the emphasis on cost rather than real-world value of products is particularly worrisome for cancer patients who are on multiple medications, dealing with chronic pain, and focusing their concern over the current care rather than the risk of addiction upon survival. Data has shown that a switch to a less expensive medication for non-medical reasons leads to increased annual medical payments. Non-medical switching has been found to be associated with significantly more negative effects than benefits, even more substantial when it occurs among patients who are stable and doing well on their current treatment regimens. The social, medical, and psychological impact arising from driving patients to addictive, yet less expensive drugs, isn’t reflected in the PBM’¬¬s profits, and it isn’t included in their formulary or pricing reviews.

How are PBMs handling concerns of persons with diabetes? Diabetes is another chronic disease that has activated discussion on the problem of PBMs, rather than the physician-guiding treatment choices. The American Diabetes Association recently published a statement Aug. 15, 2017iv “ADA Deeply Concerned with Recent Prescription Drug Formulary Trends” that patient-centered approach, in consultation with a multidisciplinary diabetes care team, should be the guiding principle for making treatment choices. The action of PBMs and insurers to limit therapeutic choice through formulary exclusion or pricing “impairs a patient’s access to certain treatments and limits provider’s choice of agents that are most effective for their patients, potentially adversely affecting patient outcomes”.

What can be done? Traditionally, physician’s prescribed medicines they felt were most likely to help their patients based on clinical judgment, rather than cost alone. We recognize that lowering pharmaceutical costs in the United States is essential. However, this must be the purview of professional societies and other health care patient and provider groups, rather than that of the PBMs that profit directly for every dollar saved. In essence, the PBM has now become the physician, without, training, knowledge, medical license, knowledge of the patient (or their illness), while making medical decisions highly relevant to the patient’s care and outcomes. Stay tuned, and it will soon be your hospital as well. The solution to the problem requires a concerted effort from those most impacted, patients and providers, to encourage conversation and collaborative problem-solving. The ultimate goal is to have the physician be the doctor once again. Without these changes, public health epidemics will continue to afflict the health and well-being of our nation.

Charles Bennett’s contact info is included in the author affiliations at the top of this page.


  1. Pitts P, Hospital Impact – PBMs are worsening the opioid epidemic. Fierce Healthcare Aug 3., 2017.
  2. Abuse Deterrent Formulations of Opioids: Effectiveness and Value. ICER June 28, 2017.
  3. Packer, M Are Payers the Leading Cause of Death in the United States? Medpage Today Sept 17, 2017.
  4. American Diabetes Association American Diabetes Association Deeply Concerned with Recent Prescription Drug Formulary Trends, Press Releases Aug. 15, 2017.

Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Using High Content Imaging for Biomarker Discovery to Support Precision Oncology

Gavin J. Gordon, MBA, PhD, Vice President of Commercial Operations at Fluidigm Corporation, South San Francisco, CA


Q: Huge new advances in technology continue to illuminate the research and translational space. How do you think your new product the Hyperion Imaging System is likely to influence precision medicine testing?

A: The Hyperion™ Imaging System* is part of a group of emerging life sciences research platforms that enable highly multiplexed immunohistochemistry (IHC). High-parameter IHC is a nascent application with the potential to drive significant advancements in precision medicine, particularly in oncology, and positively impact patient care.

The rationale for the potential impact of high-parameter IHC on precision medicine, and by extension the Hyperion Imaging System, is based on two recent observations in translational medicine.

First of all, biomarker discovery efforts in translational medicine and pharma R&D are increasingly reliant on an antibody-mediated approach for protein detection in fixed tissue sections to complement existing efforts to interrogate the genome for signatures that are predictive of drug response or that correlate with prognosis. The development of immunotherapies, primarily in oncology, is a significant driver of this trend. As a result, monitoring and characterizing the immune repertoire is an essential component of any successful proteomics-based biomarker discovery strategy for immunotherapy.

Secondly, there is significant diversity inherent in the immune repertoire that, when combined with a range of tissue-specific cellular phenotypes, requires biomarker studies that can provide “single-cell resolution” and measure a relatively large number of both cell surface and intracellular protein markers.

Both of these observations are particularly relevant for developing oncology therapy, and oncology is the most exciting area of precision medicine. Thus, a high-parameter IHC approach is particularly well-positioned to contribute to a paradigm shift in precision oncology research. However, there are relatively few options for platforms capable of meeting the requirements described above, none with simplified workflows. This has necessarily limited the scope and number of these types of studies at the current time.

Highlighting the relevance of this approach, and the pitfalls that can occur by minimizing the importance of the biomarker to the success of the drug, are recent immunology clinical trials from both BMS and Merck targeting the programmed death ligand 1 (PD-L1) pathway. Recently the BMS immunotherapy Opdivo® missed its primary endpoint in the Phase III CheckMate -026 trial, namely progression-free survival in treatment-naïve NSCLC patients whose tumors expressed PD-L1 at ≥5%. While BMS’s strategy was to include a broad subset of patients in that trial, Merck’s KEYTRUDA® clinical trial focused on a much narrower target group with higher levels of the PD-L1 biomarker. Data from the KEYNOTE-024 study showed a progression-free survival benefit in patients with tumors expressing PD-L1 on at least 50 percent of cells taking Merck’s drug. But even in the case of KEYTRUDA, not every PD-L1 positive tumor responds to the drug, and there are also responders among those who fail to meet the PD-L1 cutoff.

Clearly, single-marker IHC is insufficient to stratify cancer patients for many types of immunotherapy and predict clinical outcome, thereby justifying a higher-parameter approach to biomarker discovery for this application.

This is a problem that traditional IHC cannot easily solve, if at all, since these platforms are not capable of multiplexing at a high enough level due to inherent limitations of the antibody detection technology. Current state of the art for “high-parameter” IHC using an immunofluorescence (IF) approach enables simultaneous detection of up to 4 markers on a single tissue section, clearly not enough for biomarker discovery efforts in immuno-oncology. Specialized IF platforms can push this to 7-8 markers, but these methods are cumbersome, complex, and require significant informatics to spectrally “unmix” the signal from overlapping fluorescent emission spectrums. And it’s a lot of effort to go through since the additional data from even 7-8 markers cannot comprehensively characterize the tumor microenvironment. General consensus is that ~15-25 markers will be required to conduct the next generation of IHC-based biomarker discovery studies in immuno-oncology.

The Hyperion Imaging System, developed by Fluidigm, has overcome this insufficiency through the development of a mass cytometry (CyTOF®) platform capable of simultaneously measuring the relative expression levels of 4 to 37 proteins at single-cell resolution in fixed tissue sections [including formalin-fixed, paraffin-embedded (FFPE)] and cytological preparations. This innovative platform uses an antibody-mediated approach and metal isotopes of defined mass to enable highly multiplexed and high-throughput studies of protein expression in individual cells in situ. As a result, the Hyperion Imaging System greatly expands and builds on the inherent value of relatively low-parameter IHC (for cells on fixed tissue sections) and has the added benefit of preserving precious patient samples since data for all markers can be obtained from a single tissue section at the same time.

While it remains to be seen exactly what level of impact the Hyperion Imaging System platform will have in driving significant advancements in precision oncology, the same can be said about all other similar IHC platforms and indeed the field of high-parameter IHC in general. Time will tell as more and more studies emerge showing biological relevance and attesting to the clinical utility of such an approach.

*The Hyperion Imaging System is available only from Fluidigm Corporation. More information can be obtained at

Disclaimer: The Hyperion Imaging System is For Research Use Only. Not for use in diagnostic procedures.
Gavin Gordon’s contact info is included in the author affiliations at the top of this page.

Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Challenging Oncology Therapies With Moonshot Price Tags

Pramod John, PhD, CEO, VIVIO Health, San Leandro, CA;

Q: Some American pharmaceutical companies are well-known for pricing drugs at “whatever the market will bear”. In oncology, some specialty drugs seem to have price tags completely unrelated to the proven effectiveness of the drug. Your company has been taking a lead in confronting this problem. What do you envision as possible solutions?

A: New oncology therapies carry astronomical price tags—most people know this. Receiving far less attention is the question of actual therapeutic value. Drug manufacturers spend billions on advertisements and PR, but unfortunately, real-world patient results are frequently unimpressive. Two recent articles in BMJ make this point, 1) No evidence of benefits for popular oncology therapies and 2) Do cancer drugs improve survival or quality of life?

Why do high-cost oncology therapies with questionable results continue to be prescribed? Let’s examine a situation my company is dealing with right now. VIVIO Health received a request for neratinib, an FDA-approved extended adjuvant therapy for early-stage HER2 positive breast cancer. Our system analyzed all available performance data from sources such as the FDA, ICER and NICE. The drug approval was based on a newly created surrogate endpoint called invasive Disease-Free Survival (iDFS), which only scored 94.2% vs. 91.9% in the placebo arm. Even worse, 29% of the patients dropped out of the trial due to adverse side effects, 16.8% for diarrhea alone. Not surprisingly, the FDA committee patient representatives voted against approval.

Neratinib’s manufacturer PUMA Biotechnologies provided data on the current standard of care, trastuzumab, showing a disease-free survival (DFS) rate of 89%. Interestingly, the use of iDFS as an endpoint led to an increase in the placebo arm of ~3%, which is larger than the neratinib-to-placebo arm difference of ~2%. Ultimately the creation of a new endpoint made a larger impact than the therapy itself. The trial design itself had been altered so many times; the FDA suspected the trial had been ‘unblinded’ and attempted to determine statistically whether unblinding had occurred. Even with these highly questionable results, the FDA approved Neratinib in July.

After being shown the questionable data and asked, “Why neratinib?” the requesting oncologist explained that it’s an FDA-approved drug and “MD Anderson is giving it to everyone.”

Granted it’s hard, but physicians should have the courage to do the right thing. In the context of high-dollar, high-tech therapies and billion-dollar windfalls for pharma execs like Puma CEO Alan Auerbach, physicians must be America’s frontline ensuring that only the right therapies get to the right patients. Using Neratinib as an example, here are seven steps every physician should consider before prescribing oncology therapies:

  1. Police endpoint games. Don’t allow drug companies to define arbitrary and meaningless endpoints for your patients. Prescribe medications with objective data on meaningful endpoints such as life expectancy. Anything less should be considered experimental at best and pharma should pay for that.
  2. Do the math. In the case of Neratinib, a 2% probability of potential benefit means that for every 2 patients that might be helped, 98 are subjected to real side effects or other harm. In the neratinib trial, this equates to the ‘lucky’ 33 out of 1,420 total patients, which is quite a needle in the haystack.
  3. Consider the actual cost. Spending $5M per patient ‘helped’ with such uncertain outcomes makes no sense.
  4. Consider societal opportunity cost. Spending money on therapies that don’t work diverts dollars away from developing therapies that do.
  5. Stop listening to key opinion leaders (KOL). Dig deeper and make your own decision. A KOL’s opinion isn’t data and is too often wrought with conflict.
  6. Require companion tests. Don’t prescribe low-probability therapies without some form of a companion diagnostic and insist that the drug company provide it for you.
  7. Prescribe therapies as if you’re the patient and you’re spending your own money.

Physicians, you hold the key to changing the cost curve for ineffective therapies. Drug companies will get the message when you refuse to prescribe treatments that don’t work and cost too much.

* * *
Pramod John is CEO of VIVIO Health, a specialty drug management company providing better outcomes at lower costs.

Pramod John’s contact info is included in the author affiliations at the top of this page.

Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Accuracy and Precision Define Radiation Oncology

Eddy Yang, MD, PhD, Professor and Vice Chair of Translational Sciences Department of Radiation Oncology; Deputy Director, Associate Director of Precision Oncology at the Hugh Kaul Precision Medicine Institute; Birmingham, AL;


Q: You are a radiation oncologist with a particular interest in cancer of the prostate. How does the molecular study of prostate, as well as other cancers, including Next Generation Sequencing (NGS), help inform Precision Radiation Oncology?

A: Radiation oncology is a specialty where the accuracy and precision of treatment delivery is vital to the safety and outcomes of our patients. Many specialized techniques are utilized to enhance this precision, including intensity modulated radiation therapy, image-guided radiation therapy, and volumetric arc therapy. Emerging modalities such as proton and carbon therapy take advantage of the physics of heavy ions to potentially minimize normal tissue toxicity. With these methods, we are in essence, performing precision oncology, tailoring radiotherapy to each individual patient. However, precision oncology is much more than that, as novel technologies have expanded our understanding of the drivers of cancer that may be targetable or dictate response to treatment. Currently, emerging evidence has shown the benefits of biomarker-directed systemic treatments, but what about genomic markers to guide radiation therapy? Although the preclinical and retrospective data supports the notion of this possibility, results from prospective studies are not yet available.

Perhaps the most promising and straightforward example of biomarker-directed radiation therapy is in head and neck cancer (HNC). The human papilloma virus (HPV) has been identified as a cause of head and neck cancer. Compared to non-HPV induced HNCs, HPV associated HNCs have better outcomes due to their increased radiation sensitivity. Efforts to de-escalate radiation therapy in this population have yielded encouraging results, and prospective randomized studies are ongoing. However, more work is needed, as not all HPV-associated HNCs have good outcomes. Identifying this subset of patients using genomics and other information is crucial to optimize precision radiation therapy for these patients.

Along these lines, the prediction of tumor sensitivity to radiation would be a powerful tool for precision radiation oncology. The Radiation Sensitivity Index (RSI) is one such test. This molecular signature has been validated for colorectal, head and neck, esophageal, and breast cancers, and patients with tumors that have a radiosensitive signature had improved outcomes. Prospective trials incorporating the RSI to help inform treatment decisions are needed to validate the clinical utility of this signature.

In prostate cancer, a number of genomic signatures have also been reported that can identify more aggressive disease that is at high risk for metastasis or recurrence. Additionally, investigators recently reported the subtyping of prostate cancers using a genomic test approved in breast cancer that could potentially classify prostate cancers that respond well to androgen-deprivation therapy. Similar to the RSI, prospective validation of these findings is needed.

In summary, recent advances in tumor genomics have created potential opportunities to guide and optimize radiation treatment for cancer patients beyond traditional technical improvements. Results are promising, and prospective validation studies are currently ongoing. Given these exciting results, precision radiation oncology will be a part of the standard treatment algorithm in the near future.

Eddy Yang’s contact info is included in the author affiliations at the top of this page.

Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Precision Oncology in the UK in 2017

Justin Stebbing, MA FRCP FRCPath PhD, Professor of Cancer Medicine and Medical Oncology at the Imperial College/Imperial Healthcare NHS Trust Charing Cross Hospital, London, UK;


Q: Science is universal but the practice of medicine varies widely by geography, economics, and culture. The United Kingdom is an advanced developed country with universal healthcare. How does your country approach the issue of precision or molecular oncology?

A: Great Britain and its major cities such as London are steadily becoming one of the leading healthcare destinations in the world, for patients, their families, and the best researchers and doctors there are. The combination of individualised but multi-disciplinary care, led by our academics at universities, medical schools, collaborations with industry, and critical mass is already prolonging quality and quantity of life and increasing the cure rate for most diseases, whether they affect men or women, young or old, and are common or rare.

Within this, there is an emerging paradigm of precision cancer care in which the use of molecular data at the point of therapy directly impacts patient treatment and clinical decision-making that is joined up between the patient and their medical team. This has already had a substantial, clinical impact on many aspects of healthcare services but up until now these have been surprisingly infrequent. Times are changing now. Oncology and cancer care is now at the frontline of personalised medicine, moving beyond the previous model of giving cancer therapeutics based on trials of largely unselected patients beyond a simple phenotypic marker. We are leading the way in utilising the molecular profile of an individual’s cancer genome to optimise their disease management – to treat the right patient, at the right time, with the right tumour, with the right treatment – including combinations of medicines, surgery and targeted radiotherapy. At the centre of this is the patient, with personalised medicine offering the promise of delivering safe and efficacious cancer treatments that are targeted, biologically rational, and avoid over- and under-treatment common with traditional chemotherapy, thus reducing toxicities associated with non-specific modes of action of chemotherapy.

Advances in personalized cancer care on the medical side includes well-established molecularly targeted therapies for patients with different sub-type of cancer including HER2-positive breast cancer, BRAF-mutant metastatic melanoma, EGFR-mutant or ALK-translocated lung cancer, and  BCR–ABL-translocation-positive chronic myelogenous leukemia. Clinical trials in the UK and London have led to many advances in treating these tumour types, turning them into long-term diseases that people live with, as opposed to dying from; often after treatment, the life expectancy of that person returns to normal. In London we are introducing large-scale genomic technologies at the point of care to aim to: i) catalyse discoveries in translational oncology and drive new research that aims to dissect selective responses to targeted, combination immunotherapies and chemotherapies, ii) identify new targets for which therapies are now in development and, iii) establish in real time, without delays, the right therapy for that person. Indeed, generated by the ability to generate increasingly complex molecular data directly from patient tumor and germ-line (the underlying genetics of a person) samples, the cycle of translating discoveries into clinical practice is accelerating at supersonic speed. Using cell-free DNA, which includes DNA derived from every cell type in a person’s cancer, on a rapid ongoing basis should allow this to occur. Research at Imperial College is establishing the role of cell-free DNA in diagnosis, prognosis, and prediction in order to best treat someone, as well as follow the course of disease. Remarkably, there is an infinite requirement for this, as tumours can change by the second and no two tumour cells are exactly the same. Indeed, the stupidest cancer cell is cleverer than the cleverest oncologist. A molecular arms race to keep up would seem the safest strategy but this has been constrained for too long by cost. There are now many new discoveries in personalised and precision medicine to live a healthier, longer life but until recently even in an exceptionally well-heeled practice, most patients are not willing or able to pay enough for the doctors to devote as much attention to each individual patient, to acquire the necessary equipment, to employ sufficiently well-trained staff, and to spend the time and effort on continuing education to make such a practice viable. Going forward, the vision for world-leading healthcare in London is that each patient now would be an individual – a one size fits all approach would be relegated. This isn’t just for cancer, it applies to every specialty from primary care and paediatrics, to intensive care, endocrinology, gastroenterology, cardiology, neurology, rheumatology and all of the surgical specialties alongside which true multidisciplinary care occurs.

Clinical trials, both led by doctors and others led by biotechnology and pharmaceutical companies are moving to the forefront of care in London, for devices and techniques as well as drugs. In conjunction with NHS England, the 100,000 Genomes Project aims to bring the benefits of personalized medicine to the NHS and make this an everyday reality across diseases. To make sure patients benefit from innovations in genomics, the British Government has committed to sequencing 100,000 whole human genomes, from 70,000 patients, by the end of this year. Successful delivery of the 100,000 Genome Project will enable us to achieve a number of ambitions including: i) to be the first country in the world to sequence 100,000 whole human genomes for the treatment of patients with rare/inherited diseases or common cancers, ii) to have high consent rates from patients and public support for genomics, iii) to have established world leading genomics services within the NHS, iv) to have educated and trained health professionals within the NHS in genomics and its applications for improved patient care and treatment, raising broad awareness and understanding of the advantages genomic medicine offers to patient care in the NHS, v) to be the home of world-leading genomics companies which will work in partnership with the NHS and its academic research partners, and vi) to have stimulated the development of diagnostics, devices, medicines and treatments based on a new understanding of the genetic and molecular basis of disease. This is the ultimate in what’s called bench-to-bedside translational approaches, and we are leading the way here.

Some patients taking part in the 100,000 Genome Project will benefit because a conclusive diagnosis can be reached for a rare/inherited disease or because a “stratified” cancer treatment can be chosen that is most suitable to their individual cancer. For most, the benefit will be in knowing that they will be helping people with similar diseases in the future through research on the genome and associated clinical data they generously allow to be studied. Their involvement in the project will allow an infrastructure to be developed, which in the future will support genomic services to be applied more widely to patient care in the NHS and across many clinical specialties. This will naturally lead to benefits in the private sector because the same individuals work in both sectors in London. For example, I personally work at Imperial College where I have my research laboratory; I work at Imperial College Healthcare NHS Trust where I see NHS patients; and I have a small clinic on Harley Street where I see some private patients and am involved in a number of trials of new immunotherapy combinations. Many of my colleagues are simply the best in the world at what they do. Recently, the UK Government’s Chief Medical Officer Professor Dame Sally Davies’ report called “Generation Genome” discusses how we can offer DNA testing as standard care in so many diseases like cancer (we are already using this to identify and treat different infection strains). In my laboratory we are studying the way our genetic code is turned on or off (or regulated) and the way a cell can turn into a specific type of cell including stem cells and cancer cells. We hope to use this to benefit our patients. The combination with artificial intelligence and machine learning to help us both store and analyse multiple genomes will be critical here.

Our scientists at Imperial College, Oxford and Cambridge (a so-called ‘Golden Triangle’), the Crick, University College London, King’s are leading the way in new technologies such as CRISPR. Many human diseases are caused by the mutation of a stretch of DNA, even with one change in a single nucleotide. Adapted from a primitive bacterial immune system, CRISPR does its handiwork by first cutting the double-stranded DNA at a target site in a genome and in doing so, gives us an ability to greatly alter genetic material as a new therapy for diseases. This has been used in London to understand early embryonic development and use stem cells more reliably to treat disease. Whilst the main aim is to improve the lives of patients, there are potentially many economic benefits for the nation and ultimately the world we live in. Some may be unexpected, built on new, as yet undiscovered technologies that will emerge and we need to be patient. But such benefits may be improved diagnostic tests, better tailored treatments, and development of new treatments and medicines. The role of diet, exercise, complementary medicine utilising techniques from hyperthermia to metronomic chemotherapy, acupuncture to anti-oxidants (as per my own pieces in the Lancet Oncology on the subject of complementary medicines) would be at the forefront of care here, not relegated to an afterthought. It would place the patient’s psychologic state of wellbeing at the centre of their outcomes (in many ways this is as important as the surgery or medicines); one without the other is sub-optimal. Finally, we aim to prospectively collect data here, own those data, and link phenotype such as behaviour and history to the genetics of the cancer or any disease, and the individual as a whole. In doing so, we aim to transform the lives of our patients and their families, disrupt the course of disease, and offer the very best healthcare in the world. This is occurring in real time now, and data generated should benefit patients and their families for decades to come.

Justin Stebbing’s contact info is included in the author affiliations at the top of this page.

Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Using Data Analytics to Make an Impact in Brain Cancer

Lisandra West-Odell, PhD, Scientist and Product Manager, Cancer Commons, Los Altos, CA;


Q: Reflecting the current reality of a brain cancer diagnosis, how can Cancer Commonspositively impact the treatment and outcome of each patient diagnosed with brain cancer?

A: Just over two and a half years ago the March 30, 2015 issue of Time magazine arrived in my mailbox. The cover story “Closing the Cancer Gap” featured two women: “Both of these women have brain tumors…One of them is beating the odds”. In essence, the article compared two women diagnosed with glioblastoma and evaluated the impact of tumor genetic testing in their treatments and outcomes. The first was treated with standard of care chemotherapy. The second received the BRAF inhibitor vemurafenib in a basket trial after genetic testing identified a rare BRAF mutation in her cancer. Treatment with vemurafenib lead to unprecedented tumor regression with few side effects allowing her to beat the odds typical of glioblastoma – the most aggressive form of brain cancer.

I was gripped by this story because as a scientist (working at the time for CollabRx on the content and curation team mining treatment rationales for individual variants) it was apparent to me that treatment selection based on identification of predictive biomarkers has the potential to make a significant difference in outcomes for patients. But I saw the other side of the argument too. Comprehensive sequencing is expensive, identification of driving alterations is rare, and most often more questions are raised than answered. In my current position, scientist and product manager atCancer Commons (CC), I’d like to say a bit more about how a non-profit organization such as CC can move the needle in a landscape as bleak as brain cancer.

We must start collecting patient journeys before we understand how dozens of genetic abnormalities (or more, and their infinite combinations) propel growth, drive tumor evolution, and contribute to therapy response or evasion. I am reminded of the old adage: “Lessons come from the journey, not the destination.” Where one patient’s story can only exist as an anecdote, many stories can be organized into larger buckets or cohorts elevating them to a higher level of evidence. For this purpose, Cancer Commons is building an interactive Patient Registry that can accommodate both prospective and retrospective patient data. We are capturing a specific and limited set of data necessary to understand each patient’s diagnosis and treatment and match them to relevant treatment insights and clinical trials. We follow the patients over time to capture outcomes data so that we can inform future patients what worked and what didn’t.

Importantly, as a companion to the Patient Registry, we are building a case analytics exploratory tool. The case exploration tool allows us to display aggregate data in the registry in the form of graphs and statistics. We can thus visualize the distribution of treatments, the frequency of treatments given, and the outcomes associated with each. We can ask questions such as: Which treatments have been the most efficacious with the highest quality of life? Are any investigational drugs in trials performing well within a patient cohort? Do exceptional responders share any genetic features? What is the next best treatment option for this unique patient right now? All of this is being piloted in brain cancer as a proof of concept. But the approach of knowledge collection, analysis, and sharing will be extended to every cancer type.

I would like to invite YOU to get involved with the Cancer Commons – we are taking all comers. If you are an advocacy group, foundation, or health care organization, join our platform – we can white label our tools and services to support your patients. Share your de-identified data if you have it – the more data we have in our repository, the more powerful our analytics tool becomes. Physicians, key opinion leaders and researchers, contribute your treatment rationales and insights to collaborate on difficult cases, and help us understand how we can improve our data collection methods and interpret our findings. And finally, if you are the patient, all of this was built for you. No matter your cancer type or stage, or location or financial situation, our expert network can help you tap into the world’s collective cancer knowledge. There is a better way forward. Please come help us forge it.

Lisandra West’s contact info is included in the author affiliations at the top of this page.

Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Life After Oncology

Professor Michael Baum, Professor Emeritus of Surgery & Visiting Professor of Medical Humanities, University College, London, UK;


Q: You have recently decided to “withdraw from the field” after a distinguished multi-decade career in Surgical Oncology. Many of our readers will confront a similar choice. How do you see your life evolving from here on?

A: I qualified as a doctor in 1960 and was appointed to my first chair of surgery, at Kings College London 20 years later. In 1981, I established the first clinical trials center for cancer in the UK in my department. To succeed in that field, I needed to become cognizant of the latest teachings in moral and scientific philosophy. Ten years later I was headhunted for the chair of surgery at the Royal Marsden Hospital, the center for our National Institute of Cancer Research. Finally, I was tempted to take up an offer of a professorship at University College London in 1997, during which time I helped develop a course in “Medical Humanities”. In the UK, you are not encouraged to continue operating after the age of 70, so I relinquished my clinical chair but was kept on as visiting professor of Medical Humanities.

I continued my role in running RCTs for the treatment of solid tumors but had time to teach my students on the role of moral and scientific philosophy, history, literature, theatre and fine art, in the practice of medicine. I was also editor-in-chief of the International Journal of Surgery. I honed my skills in creative writing through the medium of my monthly editorials.

All my life I’ve loved drawing and painting, so I filled up the remainder of my time by studying painting in art schools. My ambition was to become a full-time author and artist when I eventually retired. Trouble was that the older I got, the more I became interested in the study of oncology! I then tried to combine my enthusiasm for science, art, and literature by writing provocative papers such as “Does breast cancer exist in a state of chaos?” [1] and “Why does the weeping willow weep?” [2] Eventually, as I was approaching my 80th birthday my family ganged up on me to abandon all my academic activities and long-haul trips to conferences, encouraging me to fully retire to write books and paint “great works of art”! My last trip was to deliver a talk on intraoperative radiotherapy in Las Vegas in early May and I’m happy to say that I survived that experience to celebrate my 80th on May 31st. Since then one of my art works has appeared on the cover of the Red J [3] and my second novel, “Aaron’s Rod” [4] (linked to my interests in Biblical archaeology) was published this month.

I consider myself lucky to have survived the rigors of a life in surgical oncology long enough to relaunch myself in a new career or two. I strongly recommend all oncologists to plan for their retirement, not only in financial terms, but also to maintain the health of their brain.

Professor Baum’s contact info is included in the author affiliations at the top of this page.

  1. Baum M, Chaplain M, Anderson A, Douek M, Vaidya JS. Does breast cancer exist in a state of chaos? Eur J Cancer 1999; 35: 886–91.
  2. Baum.M, Why Does the Weeping Willow Weep? Reconceptualizing Oncogenesis in Breast Cancer. N Engl J Med 373;13, September 24, 2015

Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Huge Progress in Palliative Care

Diane E. Meier, MD, FACP, Director, Center to Advance Palliative Care; Professor of Geriatrics and Palliative Medicine, Icahn School of Medicine at Mount Sinai; New York, NY;


Q: You wrote in MedGenMed in 2007 that palliative care was the job of all hospitals(MedGenMed 2007; 9(3) 6. July 7. PMCID:PMC2100088). In October 2017 you were honored at the National Academy of Medicine for your achievements in this field. How fully has your charge to hospitals in 2007 been realized?

A: Palliative care is a fairly new medical specialty devoted to reducing suffering and improving quality of life for people living with serious illness-whether the disease is curable, chronic, or life threatening and progressive. Palliative care teams work alongside disease treatment specialists to provide an added layer of support in service of pain and symptom management, family support, attention to the social determinants of health, and skilled communication about what to expect and what matters most to the patient in the context of the reality of the illness. Multiple studies demonstrate palliative care’s contribution to achievement of the triple aim: better experience of care, better care outcomes (including survival in several studies), and as an epiphenomenon of better care, much lower unnecessary utilization of 911 calls, ED visits, and hospitalization.

Until recently, palliative care was only available through hospice, a Medicare funded benefit limited by statute to people with a short (< 6 month) prognosis who agree to give up insurance coverage for treatment of their terminal illness. Not surprisingly, most people choose not to give up coverage for treatment and, as a result, the median length of stay in hospice is only 17 days with more than 30% of hospice patients receiving such care for less than a week. Hospitals are filled with patients pursuing disease treatment for one or more serious illness who are either not hospice eligible or not willing to give up treatment. The evidence of suffering- physical symptom distress, depression, anxiety, confusion about what to expect, family caregiver exhaustion- was growing in the medical literature and clinicians working in hospitals developed hospital palliative care teams to try to respond to this need. But by the year 2000, fewer than 20% of US hospitals reported any palliative care capacity. Today that number exceeds 80%- four out of five US hospitals now report a palliative care team, and among those hospitals with more than 300 beds (the tertiary and quaternary care settings that serve the sickest and most complex Americans), over 90% now have a palliative care team.

While growth in prevalence of hospital programs improves patient access (now 80% of all hospitalized patients in the US receive care in an organization with a palliative care team), access is not the same as quality. Only 39% of hospital palliative care programs meet guidelines for staffing levels and disciplines and fewer than 10% of 1,800 U.S. hospital palliative care teams have achieved the optional Joint Commission advanced certification in hospital palliative care, a marker of consistent guideline adherence and quality.

But there are challenges that go beyond the need for greater accountability for quality and standardization among hospital palliative care teams. The greatest current challenge in the field is the recognition that the great majority of people living with serious illness are neither dying (and are therefore ineligible for hospice) nor hospitalized- hence the real gap in access is in community settings including patient’s homes, nursing facilities, cancer centers, dialysis units, and office practices. The next 10 years of our organization’s work will be committed to both ensuring quality and standardization incentives and requirements for palliative care programs regardless of setting and in markedly improving access in American communities nationwide.

Diane E. Meier’s contact info is included in the author affiliations at the top of this page.

Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.