Reimbursement and Managed Care News for November 2007

Off-Label Use of Oncology Medications The List of Acceptable Evidence Based Oncology Journals Broadens

It is widely recognized that oncologists often experiment by treating cancer with medications that are not specifically approved for the neoplasm or carcinoma being managed. The Centers for Medicare and Medicaid Services acknowledges that payment for oncology medications cannot be limited solely to oncology indications for agents that are approved by the Food and Drug Administration.

In 1993, the federal government constructed a list of journals that it believed was a good source of credible, evidence-based data on evolving indications for oncology meds. However, the list had not been updated in 14 years. As a result, it recently added 14 journals to its list of sources for deemed acceptable oncology articles on off-label indications. New additions to that list, last updated in 1993, are based on recommendations of the American Society of Clinical Oncologists. The Table lists the additions to this group of peer-reviewed information outlets, which is part of chapter 15, section 50.4.5 of the Medicare Benefit Policy Manual, Publication 100-02.

Journals added to the list of acceptable publications for information on off-label information

Annals of Oncology

Biology of Blood and Marrow Transplantation

Bone Marrow Transplantation

Gynecologic Oncology

Clinical Cancer Research

International Journal of Radiation, Oncology, Biology, and Physics

Journal of NCCN

Radiation Oncology

Annals of Surgical Oncology

Journal of Urology

Lancet Oncology

Unlabeled Use for Anti-Cancer Drugs: Medical Literature Used to Determine Medically Accepted Indications for Drugs and Biologicals Used in Anti-Cancer Treatment: Transmittal 78. Centers for Medicare and Medicaid Services (www.cms.hhs.gov/transmittals/ downloads/R78BP.pdf), September 21, 2007.

More Work to Do on Breast Cancer Guideline Adoption

Medical practice variation has long been known to be a national and local problem, which results in serious quality of care gaps. Oncology practice apparently is not immune to these quality dilemmas, according to recently released data.

Guidelines for hormone receptor—positive breast cancer have been in place since 2002 from the National Comprehensive Cancer Network (NCCN), but a basic component of these guidelines is followed by just more than one half of the NCCNmember institutions, according to investigators from the Dana-Farber Cancer Institute in Boston. If this is the case in NCCN institutions, what is the situation in the average community oncologists’ practice?

The researchers evaluated the cancer care of 3,190 women diagnosed with hormone receptor—positive, node-negative breast cancer (tumor size > 1 cm) and treated at one of eight member NCCN institutions. Fifty-six percent of the women evaluated received adjuvant chemotherapy, which is advocated by the guidelines as backed by solid evidence of benefit.

A logistic regression analysis revealed that adjuvant chemotherapy was less often used in women with tumors smaller than 2 cm (48%) compared with women with larger tumors (83%). There was considerable variation in chemotherapy use from one treatment facility to another (range of guideline compliance, 47%—65%).

Data from before 2002 and after the introduction of the guideline illustrated just how variable the practices were. The investigators found that just after introduction of the guidelines, institutional response ranged from a 25% decrease in prescription of adjuvant chemotherapy to an increase of 28%.

Hassett MJ: Chemotherapy use for hormone receptor—positive, node-negative breast cancer. Presented at the 2007 Breast Cancer Symposium of the American Society of Clinical Oncology, San Francisco, September 17, 2007.

Multiple Biopsies Are Needed to Diagnose Prostate Cancer

The use of prostate-specific antigen (PSA) screening yields many false-positive (as well as false-negative) results. Ordinarily, a positive PSA test will be followed by needle biopsy of the prostate. What is the rate of cancer diagnosis by this method, and how many biopsies over the years should Medicare Advantage plans be prepared to cover in a patient in whom cancer has not been found previously?

Researchers from the Veterans Affairs system, and the Dartmouth and Harvard Medical Schools, evaluated the Surveillance, Epidemiology, and End Results (SEER) database for Medicare beneficiaries who had undergone needle biopsies of the prostate between 1993 and 2001. Of the 8,273 men who underwent 10,429 biopsies in one SEER geographic area, prostate cancer was diagnosed in 32% of the tests.

Diminishing returns: Frequency of prostate cancer diagnosis by number of sequential biopsies in study group.

Biopsies

Prostate Cancer Diagnosis

1

32%

2

50%

3

62%

4

68%

The frequency of prostate cancer diagnosis after two biopsies was 50%; after three biopsies, it increased to 62%; and after four biopsies, it rose to 68%. The investigators found that in men whose initial biopsies were negative, the incidence of a second biopsy within a year was 12% and a third biopsy within five years was 38%.

The implications of these findings are unclear: for instance, does this mean that initial biopsies yield too many false negatives? Are multiple, sequential biopsies in these older patients finding aggressive cancers or those which do not progress rapidly over time? Finally, should payers pay for multiple prostate biopsies in older patients who have in the past had negative results?

Welch HG, Fisher ES, Gottlieb DJ, et al: Detection of prostate cancer via biopsy in the Medicare— SEER population during the PSA era

2007;99:1395-1400.

. J Natl Cancer Inst

Which Model Best Predicts Which Women Will Test Positive for BRCA-1 and BRCA-2 Mutations?

Since the discovery of mutations to the BRCA-1 and BRCA-2 genes has been linked to the majority of inheritable breast cancers in the last decade, insurers and managed care plans have had the difficult task of deciding how to make these tests for these gene mutations accessible and how members with positive tests should proceed. The tests are expensive, at about $1,000 per patient, and health plans are wary about allowing each member to be tested. Furthermore, multiple tests have been developed, but they can provide different results in the same patient. For this reason, managed care is interested in predictive modeling to determine which health plan member should be allowed to obtain these tests.

Annals of Internal Medicine

A new study from multiple centers around the country tested seven models that attempt to predict if a particular patient would have a positive gene test. The authors of this study produced the results of their tests in , showing that these seven models did a fair job of predicting which patients would have positively identified as having BRCA-1 or BRCA-2 mutations. The best performing models accurately predicted the results around 80% of the time in patients with a family history of breast cancer and other risk factors. One model, BRACPRO, was slightly better than the others, but it does matter which patient subgroups are the subject of the model. In patient groups not considered high risk, however, all of the models had high false-negative and false-positive results.

The authors conclude that any of these seven models can be used in high-risk populations to help inform the decision to send specific patients for genetic testing and counseling. The implication for health plans, insurers, and payers is that they should investigate the use of these models (and restrict their use to high-risk patients) to help define which patients might be appropriate for actual testing for BRCA-1 and BRCA-2 mutations.

Parmigiani G, Chen S, Iversen E, et al: Validity of models for predicting BRCA1 and BRCA2 mutations

2007;147:441-450.

. Ann Intern Med

More Evidence in Favor of Virtual Colonoscopy

New England Journal of Medicine

Virtual colonoscopy is gaining ground as a suitable alternative to traditional colonoscopy, and at less than half the cost of the traditional procedure, this may prove to be an interesting point of contention between insurers or payers, and the gastroenterologists or radiologists performing the procedures. The newest evidence showing that virtual colonoscopy can be a reliable way of detecting advanced neoplasia comes from the .

Researchers from the University of Wisconsin, Madison, compared primary colorectal cancer screening using a type of virtual colonoscopy called computed tomographic (CT) colonography and standard optical colonoscopy in nearly 6,300 consecutive adults. The average age of the study participants was 58 years. The group was randomized to receive traditional colonoscopy or CT colonography. Patients in whom polyps were found through CT colonography were then referred for optical colonoscopy and polypectomy.

A total of 123 advanced adenomas and carcinomas was found in the CT colonography group, compared with 121 in the optical colonoscopy group (advanced neoplasia was confirmed through histology in 100 and in 107 participants, respectively). The total number of polyps removed was 561 in the CT colonography group and 2,434 in the traditional colonoscopy group. Seven patients undergoing traditional colonoscopy suffered colonic perforation compared with no patients in the CT colonography group.

The cost difference, lack of need for difficult bowel preparation, and lack of anesthesia services favors CT colonography as a general primary screening technique. Patients found to have precancerous polyps or cancerous tissue can then undergo traditional colonoscopy, according to the authors, which might be cost saving overall.

Kim DH, Pickhardt PJ, Taylor AJ, et al: CT colonography versus colonoscopy for the detection of advanced neoplasia

2007;357:1403-1412.

. N Engl J Med

National Provided Identifier

Time Is Running Out

Oncology practices that haven’t already received a national provider identification number will find their time is beginning to run out. All providers still receiving payment from the Medicare fee-for- service system must have their national provider identifier number by May 23, 2008. After that date, Medicare will no longer accept electronic claims for payment using legacy identification numbers.

The national provider identifier was mandated as part of the Health Insurance Portability and Accountability Act (HIPAA) of 1996 to assist in implementation of electronic transactions with the Centers for Medicare and Medicaid Services (CMS). The Agency began issuing national provider identifiers on May 23, 2005. After a three-year period, their use will be mandatory. According to CMS, this requirement does not apply to paper claims, as these are not “HIPAA transactions.”

Medicare Fee For Service (FFS) National Provider Identifier (NPI) Final Implementation: Transmittal 1349, Centers for Medicare and Medicaid Services (www.cms.hhs.gov/ transmittals/downloads/ R1349CP.pdf), October 5, 2007.

Pharmacogenomic Field Advances in Lung Cancer

Pharmacogenomics, the effort to match patients to optimal therapies based on their genotype, is being counted on by managed care to help define which patients should appropriately receive expensive and potentially toxic cancer chemotherapies. Through pharmacogenomics, many insurers and payers seek to blunt cost increases in the pharmacy and medical benefit associated with increased utilization of biologic and specialty pharmaceuticals.

in-vitro

One collaborative group of researchers from Duke University, Durham, North Carolina, and the H. Lee Moffitt Cancer Center, Tampa, Florida, have identified the genetic expression of non—small cell lung cancer cells. This differentiation was matched to drug sensitivity data (using tumor tissue samples from patients) to determine whether the patients would respond to or be resistant to cisplatin or pemetrexed therapy.

The authors stated that the “genomic-derived signatures of cisplatin and pemetrexed sensitivity were shown to accurately predict sensitivity in vitro” and to accurately predict treatment response in patients receiving cisplatin (Figure).

In a statement, Anil Potti, MD, of Duke University said, “We found the strongest inverse correlation between tumors that were sensitive to cisplatin and those that were sensitive to pemetrexed. This suggests that some patients [with non—small cell lung cancer] who are not likely to respond to cisplatin should perhaps be treated with pemetrexed first.”

A clinical trial is underway to confirm these findings, which may be a huge step in proving the utility and practicality of “individualized medicine.”

Hsu DS, Balakumaran BS, Acharya CR, et al: Pharmacogenomic strategies provide a rational approach to the treatment of cisplatin-resistant patients with advanced cancer

2007;25:4350-4357.

. J Clin Oncol