The Potential Utility of Studying Metastatic Disease

June 1, 2007

Patients with cancer are usually staged based on the presence of detectable regional and/or distant disease. However, staging is inexact and cM0 patients may have microscopic metastases (cM0pM1) that later cause relapse and death. Since the clinical tools used to stage patients are fairly similar for different tumors, the ratio of the rates of metachronous to synchronous metastases should be similar for different tumors (hypothesis #1). Improvements in diagnostic tools should have caused the ratio of metachronous-to-synchronous metastases to have decreased over time (hypothesis #2). Finally, the fraction of patients with either metachronous or synchronous metastases should have declined over time due to increased screening and earlier diagnoses (hypothesis #3). To test these hypotheses, Surveillance, Epidemiology, and End Results (SEER) data from 1973-1998 were analyzed for 19 solid tumors. A linear relationship was seen between the rates of metachronous and synchronous metastases, with modestly strong correlation coefficients, consistent with hypothesis #1. Over time, changes in staging methods have not significantly altered the ratio of metachronous/synchronous metastases, contrary to hypothesis #2. Also over time, a decrease in the number of patients with metastases was found, consistent with hypothesis #3. Therefore, the rate of anticipated metachronous metastases can be estimated from the rate of clinically evident metastases at presentation. Changes in screening/staging of disease over time may have reduced the overall fraction of patients with metastases.

In this issue of ONCOLOGY, Anacak et al present an interesting study examining the natural history of subclinical metachronous or synchronous metastases in patients with newly diagnosed cancer.[1] They divide patients into three basic categories-those who clinically have no overt or subclinical metastases (cM0pM0), those who have no overt metastases but later develop clinical metastases that they assume are from subclinical metastases at presentation (cM0pM1), and those with clinically evident metastases (cM1pM1).

The authors then set out three hypotheses to test in these patients. The first is that the ratio of synchronous to metachronous metastases should be similar for different tumors. The second is that improvements in diagnostic tools should have caused the ratio of synchronous to metachronous metastases to increase over time. The third is that the fraction of patients with either synchronous or metachronous lesions should have declined over time due to increased screening and earlier diagnosis. To test these hypotheses, the authors analyzed Surveillance, Epidemiology, and End Results (SEER) data from 1973 and 1998 for several different cancer types.

First Hypothesis

The bulk of the data in this report is represented in a single illustration (Figure 3), in which the rate of cM0pM1 is plotted vs the rate of cM1 for 19 different cancers from 1973 and 1998. This plot shows a modest correlation (r = ~0.5) between these two rates, which the authors believe is consistent with hypothesis number 1. However, it is not until the y axis is normalized to all patients (cM0pM1/cM0) that a strong correlation coefficient is seen (r = 0.86). The rationale for this normalization is stated, but not entirely clear.

Without this normalization, we feel that the data more likely demonstrate that the biology of these tumors is not the same-some tumors are easily identified and metastasize late (many melanomas), whereas others present with early metastases (pancreatic cancer). Therefore, there is no biologic reason to expect that the synchronous-to-metachronous rates should be the same across tumor types.

Second Hypothesis

In terms of the second hypothesis, we are not surprised that this was disproved by the data at hand. Imaging technology has dramatically improved over the past 25 years. This has led to our ability to radically improve diagnosis of both synchronous and metachronous lesions. Additionally, there have been wide-ranging improvements in the treatment of micrometastatic disease, and thus, the number of patients presenting with metachronous lesions is decreasing. This is illustrated by the SEER data in which 14 of 19 tumor types show a decrease in the rates of metachronous tumors between 1973 and 1998.

For example, there has been a steady rise in the incidence of colorectal cancer.[2] However, the incidence of synchronous and metachronous disease has decreased in this analysis of SEER data. Obviously, our ability to diagnose both of these has increased over time because of improvements in computed tomography and magnetic resonance imaging and the advent of positron-emission tomography (PET) imaging.[3] Additionally, with the advent of several new cytotoxic and biologic chemotherapeutics, the percentage of patients who develop metachronous disease has decreased substantially.[4]

One problem with SEER data in terms of identifying metachronous disease is that this characterization is represented as just as a yes/no data point in the SEER database. Presumably, finding metastatic disease is not affected by improvements in imaging, because the SEER data only show that metastases have occurred but do not address the timing of when they occurred. For tumors that spread early and for which we do not have good screening options (pancreatic and gastric cancers), we expect to see an increased incidence of synchronous disease being diagnosed. However, in tumors that spread late, improved imaging is less likely to have an impact on identifying synchronous disease.

Third Hypothesis

The data are also consistent with the third hypothesis that the fraction of patients with metastatic disease should have decreased over time. This is certainly borne out by the data presented in this article. In 16 of 19 tumor types (except pancreatic, cervical, and endometrial cancers), the fraction of synchronous metastases has decreased, and in 14 of 19 tumor types (excepting lung, liver, larynx, soft-tissue sarcoma, and endometrial cancers), the rate of metachronous lesions has decreased. Interestingly, the rates of patients presenting with synchronous metastases has fallen in almost every disease type examined despite all of the advances in imaging technology.

One might be tempted to think that the rate of synchronous metastases would increase because we are better able to diagnose them. However, as the authors correctly point out, screening techniques have also improved. For example, the incidences of both breast cancer and melanoma are substantially increasing. However, screening programs for these two diseases (mammography and skin screening) as well as prostate cancer (digital rectal exam and prostate-specific antigen assay) have dramatically reduced the numbers of patients presenting with metastatic disease (see Table 1 in Anacak et al). Unfortunately, even though screening programs are in place for breast, cervical, endometrial, and colorectal cancers, the percentage of people who take advantage of them is very small. Therefore, the benefit of screening is modest in most tumor types, and it is difficult to separate improved screening from better identification of stage IV disease.

Study Limitations

The authors readily identify several limitations to their study, one of which is that imaging studies vary for different types of cancer and changed greatly between 1973 and 1998. For example, in the three tumor types that showed an increase in the rate of synchronous disease, the use of PET scans has lagged behind that of other areas due to decreased efficacy in these tumor types. Unfortunately, it is not possible to examine the use of varying imaging techniques because these data are not available in the SEER registry.

Another drawback of studying the SEER database is the lack of consistent data regarding the use of chemotherapy and radiation therapy and the inability to determine when metastases occur (as noted above). One is unable to demonstrate that metastases are identified earlier. It is difficult to expect the incidence of developing metastases to change in this dataset unless adjuvant therapy makes a difference, and we know that in most cancers, adjuvant therapy has little impact.

Conclusions

It will be interesting to examine the utility of the data presented by Anacak et al over time. Theoretically, we could study the rates of metachronous to synchronous disease to examine the effectiveness of new systemic chemotherapy regimens. If the rate of metachronous disease decreases compared to what is expected based on the number that present with synchronous disease, one could theorize that this would be due to more effective systemic therapy. Only future studies of specific tumor types can demonstrate whether this theory will stand the test of time. We commend the authors for a very interesting and timely review of this topic.

-Adam C. Berger, MD
-Elin R. Sigurdson, MD, PhD

References:

1. Anacak Y, Meyer JJ, Marks LB: Association between the rates of synchronous and metachronous metastases: Analysis of SEER data. Oncology (Williston Park) 21:828-834, 2007.

2. Wilmink AB: Overview of the epidemiology of colorectal cancer. Dis Colon Rectum 40:483-493, 1997.

3. Herbertson R, Lee S, Tebbutt N, et al: The expanding role of PET technology in the management of patients with colorectal cancer. Ann Oncol Apr 13, 2007 (epub ahead of print).

4. Benson AB: New approaches to the adjuvant therapy of colon cancer. Oncologist 11:973-980, 2006.