Tumor-related prognostic factors for breast cancer

The number of tumor-related features available to predict the prognosis of patients with breast cancer has grown impressively in recent years. Histology, tumor stage, and lymph-node status are now supplemented with measurements of steroid hormone receptors, ploidy, S-phase fractions, growth factors, oncogenes, and oncogene products. Cellular and molecular biology have not only advanced the understanding of carcinogenesis, but have provided a host of new biologic measures potentially related to clinical outcome. Interest in prognostic factors has been stimulated by the success of systemic adjuvant therapy for early-stage, operable cancer of the breast. Any feature of a tumor, or combination of features, that accurately indicates which patients are destined for recurrence and which are not is of considerable importance. Patients destined for recurrence can be selected for systemic adjuvant therapy, while patients who will not have a recurrence can be spared the morbidity of a treatment that offers no benefit. In addition, refinement of prognostic information facilitates improved clinical testing by ensuring comparability of treatment groups and providing markers to measure the success or failure of specific therapies. The literature devoted to prognostic factors for breast cancer is extensive. Scientific reports are supplemented by a multitude of letters, reviews, and metaanalyses.1, 2 Univariate and multivariate analyses are basic techniques. Variables are individually compared with measures of outcome, and those that are significantly related to outcome are used in multivariate analyses to determine if they have independent predictive value. These are then combined to form new prognostic categories. The mix of individual variables changes, however, and interrelationships are not always consistent. Measures of outcome are multiple, and relationships to outcome are subject to change with duration of follow-up.3, 4 Confirmation of projected outcomes with prospective studies is largely lacking for newer variables.5 The complexities are such that computer models are needed for integration of information. Computerized neural networks that are designed to learn from new data and predict individual patient outcome are under development to assist clinicians in making decisions about clinical management.6-8 The purpose of this article is to review tumor-related biologic factors of current interest and relate them to prognosis and treatment objectives. Ductal carcinoma in situ (DCIS) represents a small, but important, group of preinvasive breast cancers that can almost always be cured by local-regional therapy. DCIS made up 6.3 percent of 169,260 carcinomas (4.7 percent if lobular carcinoma in situ is excluded) reported to the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) national registry between 1973–1987.9 Only one percent of these early cancers are associated with metastasis to axillary nodes, and almost all (98 percent) are cured by local-regional therapy regardless of their size. Local recurrence, particularly after breast-conserving therapy, is associated with poor nuclear grade and comedo-type necrosis, but because dissemination is infrequent, systemic adjuvant therapy is unnecessary.10, 11 Invasive carcinoma carries with it the clear potential for metastasis and a diminished opportunity for cure. The term “microinvasive” is loosely used to describe the process of invasion in its earliest beginnings and to suggest that a tumor is still highly curable, possibly as curable as noninvasive carcinoma.12 No consensus exists on the definition of such tumors, and no consensus exists for the prognosis, other than that it is favorable. Distinguishing among in situ cancers, cancers with minimal signs of invasion, and those with a few millimeters of invasion is often difficult, and judgments among pathologists are not uniform. As a group, invasive carcinomas 5 mm or less in diameter have widely divergent rates of metastasis to regional lymph nodes, ranging from three to 28 percent in various reports.13-16 These rates are far higher than for noninvasive carcinomas and imply a poorer prognosis. A few special histologic types of invasive carcinoma pose a smaller risk of dissemination and death than other types of invasive ductal carcinoma. These types are pure mucinous, pure tubular, pure medullary, and pure papillary carcinoma.17-25 Mucinous carcinomas tend to occur later in life than the other three types.9 Typical medullary carcinoma is noteworthy because of the disparity between its aggressive histologic appearance and its less aggressive clinical behavior. However, it may be more aggressive than the other types mentioned.23 These special types form a small group, representing less than six percent of all invasive carcinomas.9 Most special types are localized when diagnosed. When nodes are involved, the number of involved nodes is usually three or less. In a large study, Rosen et al26 found that patients with special histologic types measuring 1.0 cm or less in diameter a of The for all patients with measuring cm or less for of patients in the registry for of these special of invasive carcinoma are about percent compared with percent for all other types of invasive ductal A number of histologic features of invasive ductal and lobular carcinomas have prognostic when in histologic nuclear tumor as or and 28 and the of these in multivariate has been to tumor as a of tumor of can be special and is to independent of axillary grade is on the of number of and nuclear in These are combined as the grade or from to from or to or poor In are a grade is found to with tumor and with and nuclear grade are to status and tumor as prognostic features, but are of for and In a multivariate by et histologic grade of variables related to prognosis independent predictive for of and patients with after the number of axillary nodes, tumor and In a of and found that histologic and of histologic no for death rates after the number of involved axillary nodes, stage, and of tumor patients with less than 1.0 cm in et found that histologic and nuclear but not for recurrence and However, only percent of these patients histologic and nuclear grade may to adjuvant has been to a in prognosis among and patients with than among such patients with may the in outcome between and with the prognostic of nuclear and histologic are are few among the small group of histologic grade In of et found that nuclear grade a more of outcome than or not in the and nuclear grade histologic grade a with than grade only to tumor diameter as a of among to et These reported a of percent for a small group of patients with nuclear grade or with histologic grade of tumor grade is often in a but the clear of histologic and nuclear on prognosis and their with tumor and with to about their as prognostic as a prognostic is on the that of is a for tumor growth and Interest in as a prognostic stimulated by the of on tumor and by the potential for treatment with The prognostic of tumor reported by et who and in the of invasive carcinomas and found their number and significantly in with and The of metastasis with in the other have in a of of invasive breast et found considerable in in of the tumor and between the of These found no between and other tumor factors of prognostic and no between and or that in its for in the of The of a cancer or is a and of prognosis, but The localized to breast regional invasion to or metastasis to regional lymph and regional These three groups with for after and treatment but potentially curable only In of these a of are not to a from the the of localized and regional the and a of only the three and reported rates of and percent for patients The is noninvasive which are highly curable, as invasive it on of the tumor and the of the with metastasis the of invasive carcinoma and have of treatment failure and for to are and on clinical or information is to providing a of the are with biologic potential and of and a of The of are the of the tumor and the to which regional lymph nodes are These variables are but they are The of metastasis to regional nodes axillary and and to as As variables the number of and tumor more for of prognosis than of axillary lymph nodes is for the of they are large and and a poor for In a study, in percent of and in The of metastasis to axillary lymph nodes on histologic of the nodes that a tumor with the to has and may have to as analyses that the or of metastasis to axillary lymph nodes is the of recurrence and death In the of systemic adjuvant therapy, the of recurrence is percent for patients metastasis on histologic and percent for patients with nodes, a of axillary nodes, when all other axillary nodes are in only percent of and their to prognosis is of axillary metastasis is the of prognosis in operable the that a of patients axillary metastasis are not cured by local-regional therapy and patients with metastasis are and after percent and percent for patients with metastasis to or more that they are of systemic prognostic information is from axillary lymph nodes than the of The information is the number of involved The number of involved nodes a prognostic that is related to the for recurrence and related to In one large of the of patients with to 4 to and or more involved nodes and The number of nodes and by the of nodes involved, not the prognostic of the number of nodes that provided that is to all or that is is to A axillary is to The of the growth of the of the lymph nodes, and the axillary by the can be related individually to prognosis, but all are in a and tend to be a of the number of involved to of nodes in growth of and a higher axillary of growth is only with and prognosis only when three or more nodes are When the number of nodes with is the of axillary has no predictive mm in are more than and more than three found on of nodes or as on of nodes of metastasis are of interest because they potentially the of to A number of studies have found or no in of patients with compared with patients but a recent prospective by the Cancer of nodes a significantly poorer prognosis of patients with this more of in axillary nodes are in a to and of The used with and which can of related to breast predict that can one cancer among one may be or more are used to in of patients with early breast which may more to early lymph nodes are a of the breast. are not for but they are involved in percent of when no are found in axillary nodes, a group of this histologic that they are of metastasis that a tumor is to these nodes has the prognostic as metastasis to axillary However, they are less for and their small number less potential for prognosis. metastasis is only to axillary metastasis as a prognostic and it has been found more than and to axillary and nodes indicates a prognosis than metastasis to axillary nodes or nodes from about percent when group is involved to percent when are is that metastasis to nodes indicates more involved regional nodes may have the prognostic they are found only in the or the from to nodes of axillary nodes, but it can occur in the of axillary nodes or The prognosis for patients with metastasis to this is in the current with dissemination of cancer patients with metastasis their is not always as poor as that of patients with metastasis the The of the The rates for patients found with metastasis in reports and a prognosis between and The of tumor as a prognostic in of invasive carcinoma is to measurements from clinical and and histologic In analyses it is only to axillary status as independent prognostic is related to of regional number of involved axillary lymph and of recurrence and The prognosis of invasive carcinomas to and of cancers is by their smaller In one cancers to 5 mm and to mm in diameter axillary metastasis in only and percent of However, the of nodes can up to percent for of these of are they are or not and regardless of they are The of tumor on prognosis can be in and and dissemination with tumor growth of interest are tumor a available for patients and risk for status and tumor are independent on of patients with breast These variables prognostic groups with from to is for and from et 1.0 cm or less in diameter have risk of The of patients with 1.0 cm or less in diameter is to The of of invasive carcinoma measuring 1.0 cm or less on or on reported by et A large Cancer found a of Only percent of patients with 1.0 cm or less in diameter recurrence of their treatment in a by Rosen et significantly to that of patients with to cm in and it that percent of patients with cm or less cured years. from the of the or from on histologic These and other studies the that patients with the combination of and a tumor diameter of cm or less a of patients who not significantly from systemic adjuvant As failure rates are in with to axillary lymph nodes to systemic adjuvant therapy, have largely on the of other biologic variables to the prognosis of a group that is to more of and have as of prognosis and as to hormone and to percent of breast cancers of The with which and the of with patient their in of or more of are for clinical and can more than The of that for have been particularly if is is only after of its by a The clinical of to the that its to percent of patients with of in their to hormone or therapy. A higher if are and if and are Patients with have after and after recurrence compared with patients with tumors, and this is independent of axillary However, the of status as independent prognostic is diminished by its with other of prognosis and by its to hormone therapy. cancers have nuclear a S-phase a of a and a The of therapy on prognosis is to because patients and from adjuvant or hormone therapy. In studies the and of patients with are only in the of hormone the of status as a is after a of When patients not adjuvant hormone therapy percent higher for patients than However, the of the patients with one to three nodes and tumors, not hormone therapy is less of a and small but in and from to percent have been found between and after various of A multivariate of prognostic factors by that and status for more than status to be more for prognosis than tumor in but not in et found status to be less for the prognosis of or than number of nodes and nuclear The are that status is a prognostic and that status no are more of growth than of status or in combination with axillary status to a or a that has a of recurrence to systemic adjuvant therapy. and of measuring in on or on on of the number and of These with the of have been with to hormone therapy and with The to a of with a particularly is of but it indicates a more is by When found in tumors, it is in is not in other 11 are associated with in and of patients with for tumor lymph and status is still associated with early recurrence and status in patients indicates improved prognosis in and In patients with tumors, status associated with a of percent percent for In and of and In and patients of and status may be of adjuvant hormone than with of the of individual and the number of in of the is to determine the of is compared with a and to determine the of are in the or in the of the with the are in the or early and with of are in the to percent of are The are to various The of from is as the the of the by a tumor has a of of S-phase and has been a is not to measurements in to percent of tumors, and and are potential of The definition of and between and is often to of prognostic for and with from that from or is often with more and than those from it has been that be used for when percent tumor and be three than risk to with to be associated with large tumor and with nuclear is in medullary large tumors, with poor histologic tumors, and of and are associated with if Results that patients with or with a have more and than patients with or with a However, the may be In of et found a in on and only a of percent in on or percent to percent when made for tumor size. The of a of and by et patients in a large the between these variables and their to prognosis. and related because in than The percent for and percent for as or on status percent for percent for a multivariate that not a prognostic for after tumor and steroid and and patients groups with and of The of patients with and and The for patients and In smaller studies of other have to prognostic for after In a of et reported no for or to predict or to recurrence after number of nodes and tumor The for patients in the adjuvant after tumor and tumor A for patients with adjuvant and et studies of in patients and found that more predictive for treatment outcome than the studies that all found it predictive of outcome, only six of the studies found The of multivariate analyses on factors, the measure of outcome. In a of ploidy, tumor histologic and nuclear grade in et found histologic grade to be the only independent of while tumor diameter and the only independent of Patients with histologic grade a recurrence of only which not by of for of more and less Patients with with a grade and a recurrence of In a of only significantly with recurrence when tumor and status as the other In a that not found that a more of recurrence than histologic grade or tumor size. for of patients with and et reported that patients with with a percent) a of Patients with with a a poor prognosis to patients with These found no of on tumors, but later to the in of patients with small, In multivariate tumor and often as independent of prognosis in these prognostic factors to be combined to et to three prognostic groups on tumor and Only percent of patients with small with a percent) recurrence years. et found that patients with 1.0 cm or less a of percent and them in a The patients with more than 1.0 cm in diameter with less than percent or than percent of and are that may be a for less than 1.0 cm in et found that of six patients with 1.0 cm or less who with consensus group in to the clinical of in carcinoma of the breast. The group that operable breast cancers that up to a of a prognosis compared with but that the small and not to as a prognostic in multivariate analyses because of the of with more prognostic however, to have with recurrence and for and breast independent of other prognostic factors, its with tumor grade often it to as independent prognostic particularly when in less than and measure on histologic The is as the number of number of in no special but with et to with The as as number of axillary and more than tumor or histologic of to a of from to percent and from to is a measure of in of the of tumor are with and this The is then with to for one and are found in the that the are and the is as the of cancer that is used as the and in S-phase with a specific In of can be by of to of The of tumor in by these from to have found between the of and their considerable of The between and poor prognosis not to medullary biologic is not with the histologic types other than medullary is predictive of outcome and is independent of ploidy, and are associated with poor histologic and In a multivariate tumor and nuclear as of it is to tumor but not to patients with a have of to and reported that patients with cm or less in diameter that a and a of small tumors, however, is not a of not special and is often found in reports from However, has not been in largely because the is histologic is and are available more with et who have of the with found no clear of by or by for of is a that a nuclear found in in the of the and not in the is by only or be but new the on are than the of but have relationships to and are associated with poor histologic and with lymph percent as the to and et reported that of and et found a in patients nuclear is a nuclear associated with that has as a prognostic is with and and has is that is by and and by breast has and in is for is of invasion and metastasis of breast and of tend to be higher in of breast The is with or are found in one of breast of in breast cancer is associated with risk of recurrence and poor largely because of its with it has is not as have been by Most studies suggest that has for prognosis in patients with lymph patients have in studies by the In a multivariate et found that tumor and independent predictive for of early that may among patients not in a of patients the of of among may to such of with status and of to of the of and as independent is one of that have been in the process of invasion and is a that of to can various in the and can of with potential in tumor The of can be with in breast cancer that of are with a and poor As a prognostic is independent of tumor and The of has to be a for and in patients with nodes and those who are is a for in patients with In patients and in patients with tumors, have been associated with of about percent and more A number of and have been for their prognostic value. are involved with growth and and has the in breast The of is a to growth in of the et found of and to be independent markers of prognosis. one of breast cancers is not in lobular and it is more often found in the in situ of ductal carcinomas than in the invasive As a of is associated with poor prognosis, but the prognostic is almost to In the on prognosis has been and not independent of other prognostic The of as a prognostic is in the that less than of cancers As the prognostic of to who usually systemic adjuvant therapy, it the that is a for A however, has been reported in patients who not adjuvant of has been related to to adjuvant to combination and to hormone with DCIS it may be a for breast recurrence after breast-conserving therapy. in breast cancer and The a of this can in Most breast cancers the are associated with advanced stage, and recurrence, a in tumor is a tumor of which is found in to percent of breast cancers, is associated with large tumor and poor growth is to growth is a that growth and growth is the of the is in breast cancers, and or of in between its in breast cancers and prognosis have been the subject of a number of but the have been of are associated with tumors, that it less to to hormone therapy, but this has not been The tumor on the of is a of is to by in or by death nuclear has a life to be in but the by is and can be with of is the found in can be in to percent of in situ and invasive breast cancers, on the for is more in of breast cancer than in has been found in up to percent of of breast and cancer and in all of is associated with other markers of tumor poor nuclear and but it is independent of and tumor of has a on and of breast cancer In have found that testing for of to predict is to testing for tumor and et and reported for with and as about percent and In other reports the of patients has been et found that percent of the of patients nuclear for with in and from percent in to percent in those with the In this the prognostic of independent of but this has not been found by other The between and other of its to prognosis, but often it from as independent prognostic in multivariate of in the for of is among the of information about prognostic factors, which are cured and which are not outcome to be by measures of are that patients to have recurrence and to and with of to the prognostic as for clinical which to considerable These decisions are made by information that patients of recurrence small to the morbidity of adjuvant treatment than percent to the of and risk is that therapy may be features tumor and tumor to for prognosis and are information that is and are important, but are more for of hormone treatment than for prognosis. prognostic to the rates of are available and are potentially but for the their is The of consensus among about of and about and the of prospective studies about their they relate to and to other is still may is to of and that are often found in a 2 the of the information that is potentially for more this information be and are the of prognostic of patients to to a form of therapy hormone histologic grade may a higher potential for to The of may be a potential of to and hormone therapy. is that markers for or to specific will be information that have a on treatment to decisions about systemic adjuvant therapy, prognostic factors information patients can be who have of recurrence and death less than percent after potentially therapy. As adjuvant therapy can be to the risk of recurrence by the potential of in recurrence of only to three percent may not the of adjuvant therapy in such Patients with prognosis with DCIS and with axillary nodes invasive carcinomas are less than 1.0 cm in diameter or who have special histologic types of carcinoma less than cm in the other patients with number of to regional lymph nodes and patients with more than cm in diameter have recurrence rates to a from systemic therapy. patients with 1.0 to cm in diameter have prognosis with of about is in this group that measures of such as histologic or nuclear and status may have the in for or systemic The on treatment of breast cancer prognostic factors but status and tumor in no adjuvant treatment for patients with 1.0 cm or less in the and consensus prognostic groups on tumor and histologic or nuclear the for adjuvant systemic therapy of clinical that from the prognosis can be from available decisions for systemic adjuvant therapy are not on of recurrence and The and for therapy are status and the of treatment that be The and of adjuvant therapy for and often for those with the risk of treatment considerable morbidity and more than a risk of is that patients the potential and the from as they the