8 PART I
Shannon Vann, CTR, originally prepared these notes.
Any really, really small type is either 1) just wrong, 2) has been stated in different words or 3) I didn’t like it for some reason.
TAKE HOME MESSAGE:
· NON-MALIGNANT CNS TUMORS REQUIRED TO BE ACCESSIONED, ABSTRACTED AND FOLLOWED STARTING WITH DIAGNOSES MADE 1/1/2004 AND AFTER
· STANDARD SITE/HISTO DEFINITION AGREED UPON BY STANDARD-SETTING ORGANIZATIONS
· NEW RULES
o LATERALITY
o WHO GRADE
o MULTIPLE PRIMARIES
“When you wish to instruct be brief – so that people’s minds can quickly grasp what you have to say, understand your point, and retain it accurately. Unnecessary words just spill over the side of a mind already crammed to the full.”
--Cicero
Slide 4 – cover slide
Slide 5
This presentation reflects decisions and materials concerning data collection for nonmalignant CNS tumors available as of June 23, 2003, with some additions/updates in the fall of 2003.
These training materials cover data collection for benign, borderline and malignant brain tumors. Because malignant brain tumors constitute such a small percentage of overall cancer incidences, little training has been provided for this site in the past. Both the National Coordinating Council’s Brain Tumor Work Group and the NAACCR Registry Operations Committee’s Brain Tumor subcommittee agreed that this would be an ideal opportunity to provide training for data collection on all brain tumors in an effort to improve the overall quality of data on brain tumors.
Slide 6
Part I will cover
Rationale & History
Definition of Reportable Cases
Anticipated Impact on Registries
Casefinding
Slide 7
What is the rationale for benign and borderline (hereafter referred to as nonmalignant) intracranial and CNS tumor (hereafter referred to as CNS tumors) surveillance and registration?
So why should nonmalignant, as well as malignant intracranial and CNS tumors be collected?
In the early 1900’s, the pioneering neurosurgeon Harvey Cushing made the observation that some brain tumors are malignant because of their histology, and some are malignant because of their location. He meant that some tumors were not resectable and would result in the death of the patient because of mass effects on vital areas of the brain. I’d say that most of us in this room are familiar with ‘malignant histologies’, given the nature of our work – but, how about ‘malignant location’?
The brain is encased within the skull bone and the spinal cord in the vertebral column – these are unyielding structures. Thus, the appearance of a small tumor in the CNS, even a histologically benign tumor, can cause severe problems if it arises in a critical area, grows enough to obstruct CSF pathways or compress normal tissues and blood vessels to cause ischemia.
In the past 100 years, with earlier diagnosis, advances in microsurgery and radiation therapy, the maxim of Dr. Cushing still stands.
The maxim of Dr Cushing still stands after these 100 years – in spite of earlier diagnosis, and advances in microsurgery and radiation therapy:
Intracranial and/or CNS tumors, whether benign or malignant, cause clinical features/symptoms by similar mechanisms of mass effect: hemorrhage, seizure activity, and edema.
Although these tumors are individually rare, patients with benign brain tumors represent an under-appreciated financial and health burden in the United States. These cases include those tumors arising in families with an inherited tendency to develop benign and malignant brain tumors, tumors arising from developmental abnormalities, morbidity from ruptured benign brain tumors, and eventual malignant transformation in a subgroup of patients with optic nerve gliomas.
An estimated 28,600 new cases of primary malignant and benign brain tumors were diagnosed nationwide in 1995.
Approximately 12,000 people died of invasive brain tumors and 947 died of benign brain tumors during that year.
There were also 131 deaths due to tumors of uncertain behavior and 2,788 deaths due to tumors of unspecified behavior reported for those sites. (from BTWG report)
SLIDE 7 cont
Although the MCSS has been collecting ‘all brain and central nervous system neoplasms regardless of malignancy’ since 1988, nonmalignant CNS/intracranial tumors have not been systematically or consistently included in cancer surveillance systems across the U.S. so the public health community has little information to offer on either incidence or mortality patterns or trends. In addition, few analytic epidemiology research studies have been conducted to help identify risk factors for nonmalignant CNS/intracranial tumors. While these tumors are individually rare, patients with benign brain tumors represent an under-appreciated financial and health burden in the United States.
Slide 8
Prior to January 1, 2004, cancer registry data collection standard setters including the National Program of Cancer Registries (NPCR), the Surveillance, Epidemiology, and End Results (SEER) Program, and the Commission on Cancer of the American College of Surgeons (COC) required central cancer registries and COC-approved hospital cancer programs to collect data on malignant CNS tumors, but not nonmalignant CNS tumors. The MCSS has required the collection of ‘all brain and central nervous system neoplasms regardless of malignancy’ since 1988.
The incidence of malignant, benign, and borderline intracranial/CNS tumors is not large; but the clinicians treating the patients and the researchers studying the diseases felt that it was as important to study the nonmalignant CNS tumors as to study the malignant tumors. Clinicians and researchers urged the cancer surveillance community to include nonmalignant CNS tumors in cancer registries so that the data would be available for research.
In July 1992, the Central Brain Tumor Registry of the United States (CBTRUS) was established. Their mission was to report population-based incidence data on all primary CNS tumors, regardless of tumor behavior. At that time 15 state cancer registries collected benign, borderline, and malignant primary CNS tumors, and many shared aggregate data with CBTRUS (including MCSS).
The National Coordinating Council on Cancer Surveillance (NCCCS) is composed of stakeholders in the cancer surveillance community and provides an arena for identification of issues involved with cancer data collection and data use. The need for national population-based incidence data on all intracranial/CNS tumors was presented to NCCCS.
In 1996, NCCCS formed a Brain Tumor Working Group (BTWG). BTWG was asked to examine all issues involved with collection of nonmalignant CNS tumors in cancer registries.
Slide 9
About two years later, in September 1998, BTWG forwarded a report with four recommendations to NCCCS. The recommendations in their entirety are found in the BTWG report to the NCCS…your handout gives a website to visit to see the entire report.
The Council accepted recommendations 1 & 2 and deferred recommendations 3 & 4.
Slide 10
The recommendations from the NCCCS Report were:
Number 1: We recommend that the following standard definition be used for collecting precise data for all primary intracranial and CNS tumors:
The first recommendation was that a standard definition be used for collecting precise data for all primary intracranial and CNS tumors:
Primary intracranial and CNS tumors are all primary tumors occurring in the following sites, irrespective of histologic type or behavior: brain, meninges, spinal cord, cauda equina, cranial nerves and other parts of the CNS, pituitary gland, pineal gland, and craniopharyngeal duct.
(Accepted)
Slide 11
Recommendation number 2: Develop a standard site and histology definition for tabulating estimates of CNS tumors to allow comparability of information across registries. Pathologists, the North American Association of Central Cancer Registries (NAACCR), the Commission on Cancer (COC), the Surveillance, Epidemiology, & End Results (SEER) Program, the National Program of Cancer Registries (NPCR), and the International Agency for Research on Cancer (IARC) need to be involved in developing this standard. (Accepted)
Recommendation 3: All registries, hospital- and population-based, collect data for CNS tumors. This effort will necessitate a change in the COC requirements and will increase costs to the hospital-based programs. Federal funding should be allocated to supplement the additional transition and ongoing data collection costs that will be incurred by central registries. Before additional data collection is implemented, a pilot study should be conducted in multiple states to assess the procedures and quality control functions needed, as well as the costs of collecting data on these tumors. (Initially Deferred)
Slide 12
Recommendation number 4: The appropriate government and professional organizations develop and implement special training programs and curricula for central registry, hospital registry, and laboratory personnel, and develop computerized edit-checking procedures. Training for reporting and tabulating primary intracranial and CNS tumors should be offered on a regular basis. (Initially Deferred)
Slide 13
In November 2000 at the annual meeting of the Society for Neuro-Oncology, a consensus conference was convened. The group agreed with the site definition in recommendation 1 of the BTWG Report and to the development of a standard site and histology definition based on the SEER site/histology validation list. In addition, differences in the definition of CNS tumors between the surveillance community and the neuropathology clinical community were discussed. For example, the clinicians wanted lymphomas of the brain included with brain tumors. Registries have always collected brain lymphomas, but the incidence is tabulated with lymphoma not with brain. All involved recognized the importance of continuing the dialogue between the clinical community and the surveillance community.
Slide 14
In 2001 the Coordinating Council met and accepted recommendations 1 and 2 as complete. They reconvened the BTWG and asked them to work on recommendations 3 and 4.
At the request of the BTWG in January 2003, NAACCR established a Benign Brain Tumor subcommittee of their Registry Operations Committee. The subcommittee was to develop procedure guidelines needed in registry operations when nonmalignant CNS tumors were included in data collection efforts.
Concurrently, the North American Brain Tumor Coalition along with brain tumor activists brought the issue of collecting benign brain tumors to Congress.
Representative Barbara Lee (D-CA) introduced HR 239, Benign Brain Tumor Bill, “to amend the Public Health Service Act to provide for the collection of data on benign brain-related tumors through the National Program of Cancer Registries”. CDC legislative staff worked with Representative Lee and the brain tumor community to insure that the BTWG definition found in Recommendation 1 was the definition included in the legislative language of the bill. Senator Jack Reed (D-RI) drafted a senate bill. In October 2002, President Bush signed public law 107-260, the benign brain tumor cancer registries amendment act.
Slide 15
Both SEER and COC agreed to make reporting of nonmalignant brain tumors a requirement with a common implementation date for cases diagnosed January 1, 2004 and after.
Slide 16
What is the impact of collecting nonmalignant CNS tumors on registry operations?
The BTWG report estimated that the annual increase in the number of cases from adding nonmalignant CNS tumors could be calculated by doubling of the number of cases represented by the malignant CNS tumors diagnosed in a year. So, if a registry has an average of 50 malignant CNS tumors annually, the estimated increase in case load with the addition of nonmalignant CNS tumors is 50 more cases annually for a total of 100 CNS cases. This usually equates to a one percent total increased caseload for cancer registries. For hospital programs with few malignant CNS cases, the estimate of caseload increase is minimal. For those hospital programs with a large neurology service, the caseload increase may be more.
Clinicians interested in having information on benign and malignant CNS tumors may be helpful in getting additional financial support for the registry.
For CBTRUS, from 1990-1993, approximately 46 percent of the tumors reported for these sites were nonmalignant.
Fifty-one percent of the primary CNS tumors reported by MCSS from 1989-1994 were nonmalignant, and NCDB reported more than 33 percent as nonmalignant during that period.
For registries that consider making nonmalignant CNS tumors reportable,
a doubling in the overall number of CNS cases could be expected.
Slide 17
In 2002, 21 state central cancer registries collected data on all CNS tumors, regardless of behavior. In those states, the impact of collecting all CNS primary tumors may be minimal because the cases are already part of the caseload.
The increase in case load for central cancer registries adding non-malignant CNS tumors is the same as for hospitals and is estimated to be about 1% of their total annual caseload.
Slide 18
If the central cancer registry’s definition for CNS sites is not the same as the definition in the public law, the definition will have to be changed. The central registries that do not currently collect nonmalignant CNS tumors will have to make sure that their reporting law allows them to include these cases. If it does not, the state reporting law may have to be changed or amended.
Since MCSS has been collecting data on all brain and central nervous system neoplasms regardless of malignancy since 1988, we don’t need to worry about changing or amending our reporting law…but it is important to note that:
All cancer registries will use the same definition for brain-related tumors, implement data edits created for nonmalignant CNS tumors, and report rates.
Slide 19
The Benign Brain Tumors Cancer Registries Amendment Act, Public Law 107-260, refers to CNS tumors as “brain-related tumors”. Brain-related tumors are defined by public law 107-260 as follows:
“The term ‘brain-related tumor’ means a listed primary tumor (whether malignant or benign) occurring in any of the following sites: (I) The brain, meninges, spinal cord, cauda equina, a cranial nerve or nerves, or any other part of the central nervous system. (II) The pituitary gland, pineal gland, or craniopharyngeal duct.
Listed refers to listed in the ICD-O
All cancer registry standard setters have adopted this definition for reportability of intracranial/CNS tumors.
Slide 20
The ICD-O-3 codes for brain are C71.0 through C71.9. Parts of the brain are the cerebrum, frontal lobe, temporal lobe, parietal lobe, occipital lobe,
Slide 21
…the ventricle, cerebellum, brain stem, overlapping lesions of the brain, and brain, NOS
Slide 22
These sites are included in the definition of CNS tumors. The ICD-O-3 codes for meninges are C70.0 through C70.9. The ICD-O-3 code for spinal cord is C72.0, for cauda equina is C72.1.
Slide 23
Tumors of cranial nerves are also included in the definition of CNS tumors. The codes for cranial nerves are C72.2 through C72.5.
Slide 24
These other sites are also included in the definition for CNS tumors.
C72.8 is Overlapping lesion of brain and central nervous system and
C72.9 is Nervous system, NOS. The ICD-O-3 code for pituitary gland is C75.1, for craniopharyngeal duct is C75.2, and for pineal gland is C75.3.