Discussion
It is not yet clear what exactly is breast cancer. It is not clear whether it is a single disease or there are few of them having the same clinical manifestations. Also in the eventuality that there are few diseases, it is not clear if all have the same lethality potential. <> Two theories are trying to explain the evolution of breast cancer. The CONTINUUM theory state that breast cancer develops from atypical, reversible hyperplasia going through an in situ phase eventually progressing to invasive cancer. The DUAL theory on the other hand, state that invasive cancer, the only true lethal form of the disease, arises de novo without going through an in situ phase. The theory also states that this form of cancer invades rapidly and metastasizes early. <> The only chance for cure or life prolongation was proved to be early detection. Therefore, screening mammography programs were implemented in order to detect breast cancer in its early stages, before it can be clinically detected and before it has a chance to metastasize. This allowed as already proved, early treatment and relative control of the disease. The results gathered up to now, showed that most of the screening programs reduced mortality and postponed death from the disease. <> The ability to find a cancer in its early stages is dependent on objective as well as subjective factors. Among the objective factors might be enumerated – quality of the equipment, positioning technique, breast density, frequency of screening and the rate of growth of the tumor itself. The most important subjective factors are knowledge and experience. <<>> In spite of all efforts, there are still cancers eluding detection. They are named interval cancers and they are discovered subsequent to a negative screening. There are no references to the length of the screening interval or to the cause that lead to the delayed detection. <> Interval cancers might be divided into a number of categories. (1) True negative interval cancer – no sign of disease may be detected on previous screening mammogram. (2) Interpreted as benign interval cancer – a lesion that proves to be malignant showed benign morphological characteristics on the previous mammogram. (3) Retrospectively visible interval cancer – a now known lesion is seen on the previous screen mammogram. (4) Single reader interval cancer – a second reader would have discovered the lesion. (5) Technical failure interval cancer – a technically poor image hampered the reader to discover the abnormality. <> “True” and “interpreted as benign interval cancers” are mostly high grade and rapidly evolving forms of the disease, being particularly aggressive in the younger age patient group. It was claimed that they are a biological different type from the very beginning, appearing or evolving within the length of a screening interval. This assertion probably stood at the fundament of the dual theory of breast cancer but so far no biological unique characteristics were found to differentiate between the two assumed types of cancer and validate the theory. Interval cancers are advancing faster and this might be sustained by an intrinsic characteristic of the tumor itself, but also by a certain biological context of the host. A very new research, claims that clinically and regarding the aggressiveness of the tumor, there are most probably two distinct forms of breast cancer, having different estrogen receptors (ER) and progesterone receptors (PR) expression as well as different combined ER/PR expression. This imply that there are two separate forms of breast cancer, each one progressing through a multistage process. (J Clin Oncol 2001;19:18-27) It was shown that regarding size, lymph node status, vascular invasion and prognosis, interval cancer tumors tend to be intermediate between screen detected and symptomatic cancers. <> Frequency of mammographic examinations was a very disputed issue in the design of screening trials. Intervals, ranging from 12 to 36 months were used by different trials. Gathering of data in the years following the trials showed that the incidence of interval cancers is very high in the first and second years following the initial screening mammogram. Its incidence decreases abruptly in the third and later years. Roughly, about 30% of interval cancers are discovered in the first year, 45% in the second one, about 20% in the third year and the rest in the fourth and later years. The mean interval of detection was 14 month, for women in the younger age group, and 22 month for women in the older age group. Therefore in the first 3 years following the first mammogram, the longer the screening interval the higher the incidence of interval cancers. <> The frequency of screening should be established according to the rate of tumor growth. The average doubling time of a breast cancer is between 100 to 180 days. This result in a five-year period while a cancer may grow from microscopic level before it might be detected. Following this five year period, the cancer enters a time slice that is called “sojourn time”, a period in which the lesion might be detected by mammogram but not yet by clinical examination. The length of the “sojourn time” is a characteristic of each tumor type. Low grade tumors have a long “sojourn time” and good prognosis, while high grade tumors have a short “sojourn time” and poor prognosis. Unfortunately many breast tumors display intratumoral histologic heterogeneity with the poorly differentiated component outgrowing the well differentiated one. This might partially explain the second category of interval cancers that initially are slow growing, long “sojourn time”, benign looking tumors commuting into fast growing, short “sojourn time” tumors. <> Breast cancers of different histologic types have different rates of growth, arriving at the threshold of detection at different time intervals. Furthermore, according to several studies, the same histologic type tumor has a shorter “sojourn time” in women within the 40-49 years old group, accounting for more interval cancers in this group, while compared with the 50-75 years old group. <> In order to be effective, screening intervals should be half of the “sojourn time”. Long screening intervals with an abundance of high grade interval tumors might lead to the wrong conclusion that screening detects only low grade cancers that have a better prognosis. <<>> Establishing a proper screening interval might probably reduce the number of “true interval cancers” as well as the number of “interpreted as benign breast cancers”. “Retrospectively visible interval cancers” may be prevented by continuous training and by increasing the experience of the physicians involved in screening, knowing that perception and interpretation problems account for many preventable missed cancers. “Single reader interval cancers” may be precluded by establishing a second reader routine, either human or computerized. “Technical failure interval cancers” will be reduced by thorough quality assurance tests as well as by proper and continuous training of the technologists involved in performing mammograms. <> No effort should be sparred, in order to reduce the false negative rate of mammography. A second reader routine as well as strict implementation of film and mammographic techniques might reduce the number of false negative mammograms to about 50% of the actual rate. Replacing 2-3 year screening to shorter, properly established intervals, another 75% of the remaining cancers might also be theoretically reduced. This way, of 100 true and missed interval cancers, only 13 will be left as unavoidable, for the time being.
