As day two of the annual COA conference started, there were still a lot of topics to be covered by the faculty. I began in the clinical track with an interesting lecture on Lynch Syndrome.
Do You Know It When You See It?
Larry Geier, MD
Director, Clinical Cancer Genetics
Kansas City Cancer Center/University of Kansas Cancer Center
Dr. Geier set the stage by first discussing relatively common hereditary cancer syndromes:
- Breast, ovary, others
- Colon, upper GI, others
What are the hallmarks of hereditary cancers? Some of the more common signals include: family clustering, younger age at diagnosis, multiple cancers in the same person, typical phenotypes, and tumor suppressor genes.
The nice (maybe “nice” is the wrong word) thing about hereditary cancers is that they may be more preventable. We can identify at-risk patients, and then attempt to prevent the cancer. For example, 5-10% of breast cancers are hereditary and 100% of those cancers are preventable. Another example is colorectal cancer where 5% of CRC cancers are hereditary and 80-90% of those cancers are preventable.
Importantly, it is becoming increasingly recognized that genetic evaluation, when appropriate, should be done soon after diagnosis in order to optimally manage the patients who have genetically based cancer.
Dr. Geier spent some time discussing Lynch Syndrome as it relates to management of hereditary cancers. Lynch Syndrome is caused by an inherited defect in any one of several “mismatch repair” genes (MMR). Five genes are currently available for clinical testing:
These MMR genes normally correct DNA mismatch mutations that occur during DNA replication. Cancer occurs when a sufficient number of these mutations occur in critical genes – it’s just a matter of time.
The problem is that Lynch Syndrome creates genomic instability, which greatly accelerates the timeline from colon polyp to CRC for example. Instead of the usual 7-10 years, it may be only 1-3 years from polyp to CRC. As you can tell, a clean colonoscopy every 10 years is not sufficient to catch CRC in these people, and frequent screening is very important in this group of people.
So how do we diagnose Lynch Syndrome:
Ideally, physicians and other care givers will screen for Lynch Syndrome. However, traditional dependence on providers to identify these patients has been largely ineffective, probably for the following reasons:
There is a wide spectrum of cancers and physicians
Providers underestimate the prevalence of these syndromes
Too much reliance on the “slam dunk” family history
Too little attention to the phenotype that is typical for Lynch colon cancer
The bottom line is that these syndromes are more common than most realize, and they are easily missed. Physicians need to pay attention to family history of cancers other than CRC such as uterus, ovary, gastric, pancreas, and urothelial cancers. There should be universal IHC screening for MMR proteins on all CRC and endometrial cancer patients. Genetic testing should be considered soon after diagnosis, and along those lines rigorous application of “red flags” will capture the majority of families.
Integrating Palliative Care and Oncology
Rebecca Bechhold, MD, Oncology Hematology Care
Dr. Bechhold describes integrated care as palliative care + oncology care, and integrated care includes: 1) an extensive discussion of the goals of care and 2) superb symptom management.
Did you know that the US spends 2X more than any other country on cancer care, and that 20-30% of $55 billion in cancer care is spent in the last 6 months of life, without any evidence of improvement in quality of life? How about this one – did you know that a cancer diagnosis is the most common cause of personal bankruptcy in the US, regardless of insurance coverage?
There is a need for integrated care because the survival time in many cancers is so limited in late stage disease. We have also seen that palliative chemotherapy frequently causes hospitalizations, and that any end of life conversation with a doctor has been shown to lower costs in the last week of life. For these reasons, and because patients express feelings of abandonment when oncologists are not part of their end of life care, there is an established need for integrated care.
The need for integrated care is there, but how do you make it part of your practice? Consistent with many of the discussions here at the meeting, working with the payer is beneficial to everyone. For example, palliative care can decrease utilization, increase profits, increase outcomes, increase bonus payments, and reduce avoidable admissions. These are all things that should get the attention of payers.
The point is that end of life care is complex, time consuming, and deserves our best effort. But also we, as physicians, deserve adequate compensation and patient management should be paid for. In the end excellent patient management leads to increased patient and family satisfaction, decreased hospital admissions, and increased QOL for patients.
Biomarkers in Practice: A Review of Outcomes and Reimbursement Issues
Ali McBride, PharmD
Clinical Coordinator Hematology/Oncology
The University of Arizona
The first portion of Ms. McBride’s talk gave us a background on biomarkers, including discovery and development of biomarker, definitions, current applications of tumor biomarkers, and limitations of current biomarkers.
But it was when she started discussing outcomes with biomarkers that the puzzle pieces started to fit together. There are, of course, great examples of clinically successful biomarkers in CML, melanoma, and colorectal cancer for example. Ms. McBride chose CRC as a case study, demonstrating why Kras is a good biomarker for CRC patients while EGFR isn’t. As a case study, the Kras example points to the goals of personalized therapies:
All that is well and good, as long as there is reimbursement to make biomarker identification and use economically viable. As Ms. McBride transitioned into the business end, it was easy to see that one needs the other for the concept to work well in practice.
In theory payers would like biomarkers because there is less risk for them:
But of course there are concerns. Development of these molecular tests is not validated in many instances, and without that there are other problems such as clinical trial development and of course reimbursement.
To give you an idea of the role of a biomarker in cancer management, nobody argues that the ALK test should not be given and reimbursed in NSCLC patients. The widely reported price of the companion diagnostic test kit for crizotinib, developed by Abbott Molecular, is less that $200, while established reimbursement is $400 - $500. In contrast the reported price of crizotinib is $9,600 per month.
Meanwhile the FDA expects that companion tests will usually be lab tests that are co-developed with the medication, and approved at the same time. Regarding reimbursement however, if the product label specifies that biomarker testing is required to identify appropriate patients, then it must be ensured that the label reads broad enough to accommodate all current and future methods of testing for the biomarker. The problem is elucidated with the Herceptin example, when some patients had coverage denied because the laboratories contracted to test for HER2 did not have access to the HercepTest or the laboratory used different IHC testing.
Suddenly even clinically robust personalized therapies are thrown into question because reimbursement is not as straight-forward as one would expect. If the test is not reimbursed by CMS or commercial insurers, market adoption of the new therapy will be impacted. Historically, national coverage decisions have helped establish reimbursement policy for biomarkers.
CMS determines the national payment limits for 65 genetic tests described by Tier code.
Ms. Mcbride went on to describe examples where reimbursement was higher for an FDA approved Kras test, and an example where the biomarker test under water. In summary, everybody wants personalized cancer therapy, and the scientific advances have been excited and meaningful, but regulation and reimbursement of these tests is still problematic.
By Don Sharpe
You must be logged in to post a comment