The authors of “How Long Have I Had My Cancer, Doctor?” have addressed a question often contemplated by patients when they receive a diagnosis of cancer. In this article, Patrone et al estimated the average “ages” of breast, rectal, and lung carcinomas based on a modified version of Collins’ law and a review of selected medical studies that measured time to local/regional recurrence (LRF) following surgical resection.
Collins’ report,[1] published in 1956, studied the time interval between surgery and LRF in patients with Wilms tumor to calculate “average” growth rates for individual patients. Assuming the growth rate of the tumor was constant, Collins then could estimate the tumor’s age. In the 1950s, surgical biopsy techniques and histologic criteria (such as tumor grade) were routinely utilized for the diagnosis and classification of cancer. However, routine, standardized assessment of the cells within solid tumors was uncommon. Over the past few decades, through the advancement of immunological and molecular methodologies, we have developed the ability to routinely assess solid tumors and classify them into various subtypes. Cells within each tumor subtype exhibit distinct biological characteristics, such as degree of differentiation, rate of proliferation, and invasiveness. While Collins’ model can accurately estimate an average natural age for breast, rectal, or lung carcinomas, it does not take into consideration the vastly differing proliferation rates reported for the various subtypes of these tumors. For example, breast tumors classified as estrogen receptor (ER)-negative typically have much higher proliferative indices, as assessed by Ki-67 staining, than do ER-positive tumors.[2] Non–small-cell lung carcinoma encompasses several different subtypes, including adenocarcinoma, squamous cell carcinoma, large cell carcinoma, and bronchioalveolar carcinoma. Non-bronchioalveolar adenocarcinomas have lower Ki-67 scores than squamous cell carcinomas or large-cell undifferentiated carcinomas; bronchioalveolar carcinomas exhibit lower Ki-67 staining than any other histologic subtype of non–small-cell lung carcinoma.[3] Theoretically, if patient A has a tumor comprised of cells with a high proliferation rate and patient B has a tumor comprised of cells with a low proliferation rate, assuming a constant growth rate over a given period of time, patient A’s tumor will be larger than patient B’s at the end of that period. Thus, it is almost impossible to accurately estimate or compare the natural histories of the tumors of patients A and B.
Recent advances in molecular biology have enabled scientists to further characterize individual tumor cells within specific subtypes. It is now generally accepted that solid tumors are comprised of heterogeneous populations of cells with varying growth rates, clonogenicity, and invasive capabilities. Recent evidence suggests that only a small fraction of cells within tumors have the ability to initiate tumor growth.[4] Solid tumors typically have extended periods of “dormancy” that persist for many years before detectable tumors eventually arise.[5] The tumor dormancy hypothesis proposes that during the preclinical phase of tumor progression, micrometastatic tumor cells do not grow for prolonged intervals of time, and that tumor formation is a highly selective, multistep process.[6] The ability of dormant cells to escape cell cycle arrest involves the acquisition of specific genetic changes, autocrine and paracrine microenvironmental influences, and the immune response of the patient.[7] Cancer dormancy might explain why some patients relapse many years (and even decades) after their initial diagnosis and primary tumor resection. Given our current knowledge regarding solid tumor growth, it is difficult to assess whether a “linear growth model,” such as the one the authors employed, adequately represents the highly complex process involved in cell cycle transition, cellular proliferation, and disease progression.
Finally, the authors’ model does not consider systemic micrometastatic residual disease. Standard histopathologic analysis and high-resolution imaging technologies usually do not detect the early dissemination of tumor cells to hematogenous (blood and bone marrow) compartments. However, early dissemination has been reported for a variety of solid tumors, including breast, colorectal, prostate, and lung carcinomas.[8] Recent reports indicate that approximately 30% of patients with early-stage breast cancer, even those with T1/T2 disease, may harbor circulating tumor cells (CTCs) in their blood and/or disseminated tumor cells (DTCs) in their bone marrow prior to tumor resection.[9] The presence of DTCs is an independent predictor of both disease recurrence and overall survival for patients with early-stage breast cancer.[10] Similarly, DTCs have been identified in 22% to 55% of patients with lung cancer, [11] and CTCs have been identified in approximately 30% of patients with rectal carcinoma prior to the initiation of preoperative radiation and chemotherapy.[12] Systemic dissemination can occur early in disease progression, irrespective of tumor characteristics and lymph node involvement, and it influences outcomes in a significant number of cancer patients with solid tumors. Occult tumor cells such as DTCs have been described as having stem cell–like characteristics with little or low Ki-67 activity,[13] which might also explain the prolonged time to recurrence seen in many patients.
The age of a cancer patient’s disease may seem like a straightforward question. However, the natural history of a given patient’s tumor in the context of the evolution of his or her individual disease is complex. It involves not only residual tumor volume following surgical resection, but also individual tumor cell characteristics, microenvironmental factors, the immune response, and the possibility of early systemic dissemination of micrometastatic cells. While a linear model of cell growth allows clinicians to estimate the average ages of breast, rectal, and lung cancers, it does not fully encompass the complexity of an individual patient’s tumor progression.
Financial Disclosure: The authors have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.
