GIST Support International requested research pathologist Anette Duensing, MD to explain how tumor grading applies to gastrointestinal stromal tumor (GIST).

What is meant by "grading" of GIST?

Anette Duensing, MD

Grading compares tumors of the same entity and sorts them according to their aggressiveness -their "grade of malignancy". In this system, so-called "high-grade" tumors are tumors which pursue an aggressive clinical course, whereas "low-grade" tumors are the least aggressive ones.

Generally, there are two factors that are important for the determination of a tumor's grade. The first one is the "degree of differentiation". This term describes the extent to which the cancer cells are similar to healthy cells of the same kind - a concept that is based on the notion that a tumor is the least aggressive the more its cells retain the characteristics of their normal counterparts. Technically, this is done by the pathologist who compares the tumor cells to normal cells of the same tissue under the microscope, e.g. colon cancer cells and the cells lining the normal large bowel. Of course, this assessment can only be done if the originating cell type of a certain neoplasm is known. However, it can be difficult even for tumors where this relationship has been identified. Gastrointestinal stromal tumors (GISTs), for example, are thought to originate from the interstitial cells of Cajal (ICC). But ICCs, which are located in the muscular layer of the bowel wall, are relatively scant, mingle very well with the surrounding tissue and are therefore hard to detect using routine microscopy.

The second factor which adds into the equation is the tumor's degree of proliferation. The term "proliferation" describes how rapidly the tumor cells are duplicating (or dividing), and consequently how fast a tumor is growing. It is generally assumed that a tumor is the more aggressive the more cells are duplicating at a time. Therefore, the degree of proliferation is often assessed by the "mitotic count" of a tumor. Mitotic figures or mitoses are dividing cells which can be identified under the microscope at high magnification. Usually, an area comprising of a certain number of high-power fields (which is the view of a microscopic field at high magnification) is scanned for mitoses by the pathologist. The more mitotic figures are counted in each high-power field, the more cells are dividing, and hence, the more rapidly the tumor is growing.

A combination of these two assessments - the degree of differentiation and proliferation - is used to determine the grade of a certain tumor. This is typically described by four degrees of severity: Grades 1, 2, 3, and 4. The cells of Grade 1 tumors are well differentiated. These are low-grade tumors, which are considered the least aggressive in behavior. Intermediate-grade tumor cells (Grade 2) are moderately-well differentiated and believed to show a clinical behavior that is in between those of low- and high-grade tumors. Conversely, the cells of Grade 3 or Grade 4 tumors are poorly differentiated or undifferentiated, respectively. These so-called high-grade tumors are thought to exhibit the most aggressive behavior. Criteria for the individual grades vary with each form of cancer. Nevertheless, the employment of a grading system for a certain tumor type only proves useful when there is a correlation between the microscopic appearance of a neoplasm and its biologic behavior. Unfortunately, this relationship is often less than perfect, and grading has generally proved of less clinical value than has staging. The staging of cancers on the other hand is based on the size of the primary tumor, its extent of spread into the surrounding tissue and to regional lymph nodes, and the presence or absence of distant, blood-borne metastases.

What about a grading system for GISTs then?

GISTs belong to the tumor types where there really is no consensus as to whether and how to grade GISTs. Although a couple of grading systems for GISTs have been published, this does not mean that these are (i) used by all pathologists and/or (ii) that it makes sense to use them. This is not without a reason. As stated above, a grading system for a certain tumor entity can only be of relevance when the microscopic appearance of a tumor reflects its clinical course. And this is exactly the problem with GISTs. These tumors do not "stick to the rules". Whereas with certain other cancers a doctor can be quite sure that a low-grade tumor will actually follow a benign clinical course and will not metastasize or grow back in the future, this is not the case with GISTs. From what is known today, even the microscopically benign looking GIST with a low mitotic count might metastasize at some point, and there is currently no means to accurately predict the behavior of a GIST on a case to case basis. Dr. Christopher Fletcher from the Brigham and Women's Hospital, Boston, has recently published guidelines for diagnosing GISTs (Human Pathology 2002; 33:459-465). In this article he states:

Although some experts disagree and maintain that at least some benign lesions can be accurately identified on morphologic grounds alone, it seems most prudent to develop a scheme based on risk assessment, rather than try to define strict criteria to separate benign from malignant (which appears to be a practical impossibility at the current time). Thus, instead, one would indicate for a given GIST its risk (low, intermediate, or high) of pursuing an aggressive clinical course, recognizing that some subset of cases (hopefully <10%) will behave in an unpredicted or unexpected fashion and that no lesion can be definitively labeled as benign." Cited from: Fletcher CDM, Berman JJ, Corless C, Gorstein F, Lasota J, Longley BJ, Miettinen M, O'Leary TJ, Remotti H, Rubin BP, Shmookler B, Sobin SH, Weiss SW. Diagnosis of gastrointestinal stromal tumors: a consensus approach. Human Pathology 2002; 33:459-465.

Subsequently, he gives a proposed approach for defining this risk based on the overall size of the tumor and its proliferation rate (mitotic count). This approach is based on the following thoughts concerning general tumor biology: As mentioned above, the mitotic count of a tumor indicates how rapidly its cells are dividing and therefore how fast the tumor is growing. The tumor size, in combination with the proliferation rate, may be a measure of for how long a tumor may have been growing. So, theoretically at least, a relatively large tumor with a low mitotic count might have been growing slowly for a long time. On the other hand, a tumor with a very high mitotic count is more rapidly growing, achieving the same size as a slow growing tumor within a much smaller amount of time. Hence, this tumor is more aggressive and might be more prone to future metastatic or locally aggressive growth. However, this new risk assessment system for GISTs has to be validated. This means that a large number of cases will have to be classified according to this approach, and the data will have to be compared to the actual clinical course of these tumors. Finally, if a correlation between "risk of aggressive course" and actual "clinical behavior" is verified by statistical methods, this method would be ready for every day use.

Until then, unfortunately, we will have to live with the mere approximations that are currently available. After all, we are still at the beginning of understanding these particular tumors.

About the Author

Dr. Anette Duensing studied medicine in Germany and the U.K., and received her M.D. from the University of Hannover Medical School, Germany. She wrote her dissertation on molecular genetics of liver tumors in the Department of Pathology at the University of Hannover Medical School. Dr. Duensing then completed three years of residency training in this department, where she participated mainly in the morphological and molecular diagnosis of hematological diseases. In 1999, Dr. Duensing joined Dr. Jonathan Fletcher's laboratory at the Brigham and Women's Hospital in Boston for a post-doctoral fellowship. In the past few years, she has been leading the GIST-related studies in Dr. Fletcher's lab. During that time, most of her research was focused on activated intracellular signaling pathways in GISTs that are triggered by the mutant KIT protein. This research has led to the identification of proteins which might be targeted by new kinase inhibitors as a complement to KIT inhibition. In 2003, Dr. Duensing took a position as Research Assistant Professor in the Department of Pathology at the University of Pittsburgh Medical School, where she continues her research.