Bone Micrometastases are Common in Esophagogastric Cancer
May 1, 1999
Bone Micrometastases are Common in Esophagogastric Cancer
abstract & commentary
Synopsis: For most nonhematological tumors, the detection of micrometastatic disease in the bone marrow is associated with reduced survival Most patients undergoing resection of esophagogastric cancers have demonstrable micrometastatic disease at the time of initial surgery. The poor long-term survival may be related to early disease spread.
Source: O’Sullivan GC, et al. Gastroenterology 1999;116:543-548.
The appearance of malignant cells in the marrow of patients with various carcinomas, including breast and lung cancers, is widely appreciated to impart a negative prognosis.1,2 It is also a commonly held perception that tumors in the gastrointestinal tract spread to bone less commonly than many other carcinomas. To examine this perception, O’Sullivan and colleagues prospectively examined bone marrow in patients undergoing resection of esophagogastric cancers. They inspected for the following:
• prevalence of micrometastases in rib marrow;
• comparative detection rates in rib and iliac crest marrow;
• the effect of neoadjuvant therapy on detection rates; and
• viability, in vitro growth characteristics, and tumorigenicity of carcinoma cells found in bone.
For this, 50 consecutive patients with newly diagnosed and potentially resectable (for cure) esophagogastric cancer were enrolled. The histological types included 19 squamous carcinomas and 31 adenocarcinomas. Patients with known metastases demonstrable by conventional preoperative diagnostic tests were excluded. Neoadjuvant therapy was administered to 15 of these patients (5 received chemotherapy alone and 10 received combined chemotherapy and radiation therapy). An additional 23 patients (with no evidence of esophagogastric cancer) undergoing thoracotomy provided control bone marrow specimens.
At the time of surgery, a portion of rib was removed and sent to the laboratory for flow cytometry and immunohistochemistry. Carcinoma cells were detected by examining for the presence of cells expressing cell-wall cytokeratin. In a further subset of patients, tumor cells were cultured from the marrow and later injected subcutaneously in immunodeficient mice to assess tumorigenicity. Bilateral iliac crest marrow was also obtained in 27 patients and subjected to the same evaluation.
Micrometastases were detected in rib marrow in 44 of 50 patients (88%). However, analysis of iliac crest marrows detected cytokeratin-staining cells in only 15% (4 of 27 patients) compared to 89% (24 of 27) of rib marrows in the same subset of patients. Detection rates were independent of histological type or nodal status and were similar in patients with or without prior neoadjuvant therapy. In the seven patients for whom cultures were attempted, tumor cells grew in five and in all of these, the cells were tumorigenic in immunodeficient mice.
COMMENTARY
Esophagogastric cancers are among the most lethal malignancies. The vast majority of patients who present with symptoms, despite adequate resection of the primary lesion, will die of metastatic disease.3 Thus, it should come as no surprise that micrometastases are present at the time of resection. However, the high rate of discovery in the bone marrow might not have been predicted. It is likely in these patients that metastatic cells were present in other organs (liver, lung, nodes) that become clinically more apparent with time.
Nonetheless, examining rib marrow for micrometastases might be of clinical value for the management of these patients. The techniques used are straightforward and well within the capabilities of most clinical pathology laboratories. Further studies will be needed to establish whether the small subset of patients with negative marrows will have better survival or whether a larger percentage of patients will have negative marrows as screening and early detection uncovers asymptomatic patients with smaller primary tumors. If so, appropriate adjuvant treatment strategies might be developed for those with or without micrometastases. For example, those with demonstrable bone marrow metastases might be treated with systemic therapy earlier, and those without metastases might have therapy directed at reducing local recurrence.
Another important clinical point is that prior (neoadjuvant) treatment did not reduce the rate of detection of rib micrometastases. This could be for a number of reasons, but it certainly raises the question of drug resistance in metastatic cells and might explain the failure of neoadjuvant drug therapy to control the growth of metastatic lesions. O’Sullivan et al did not include in this report whether the neoadjuvant therapy resulted in primary tumor shrinkage in the same patients in whom micrometastases were observed but response of the primary tumor would be expected. In fact, a difference in the responsiveness of the primary tumor and the metastases would support the notion that the metastatic cells have the added advantage of acquired drug resistance.
From a biological perspective, another interesting question is the differential detection rate between iliac crest and rib marrow. It is possible that the explanation is technical. Iliac crest marrow aspirations are more likely to be contaminated with peripheral blood than marrow teased from a surgically resected rib. Further investigation should clarify this point.
Questions raised by this intriguing and important report may result in new clinical research strategies to assess the efficacy of systemic therapy in the management of esophagogastric, and perhaps, other cancers.
References
1. Mansi JL, et al. Br Med J 1987;295:1093-1096.
2. Cote RJ, et al. J Clin Oncol 1991;9:1749-1756.
3. Blot WJ. Semin Oncol 1994;21:403-410.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content