Anesthesia and Circulating Tumor Cells in Primary Breast Cancer Patients

A Randomized Controlled Trial

Frédérique Hovaguimian, M.D.; Julia Braun, Ph.D.; Birgit Roth Z'graggen, Ph.D.; Martin Schläpfer, M.D.; Claudia Dumrese, Ph.D.; Christina Ewald, Ph.D.; Konstantin J. Dedes, M.D.; Daniel Fink, M.D.; Urs Rölli, M.Sc.; Manfred Seeberger, M.D.; Christoph Tausch, M.D.; Bärbel Papassotiropoulos, M.D.; Milo A. Puhan, Ph.D.; Beatrice Beck-Schimmer, M.D.


Anesthesiology. 2020;133(3):548-558. 

In This Article

Abstract and Introduction


Background: The effect of anesthetic drugs on cancer outcomes remains unclear. This trial aimed to assess postoperative circulating tumor cell counts-an independent prognostic factor for breast cancer-to determine how anesthesia may indirectly affect prognosis. It was hypothesized that patients receiving sevoflurane would have higher postoperative tumor cell counts.

Methods: The parallel, randomized controlled trial was conducted in two centers in Switzerland. Patients aged 18 to 85 yr without metastases and scheduled for primary breast cancer surgery were eligible. The patients were randomly assigned to either sevoflurane or propofol anesthesia. The patients and outcome assessors were blinded. The primary outcome was circulating tumor cell counts over time, assessed at three time points postoperatively (0, 48, and 72 h) by the CellSearch assay. Secondary outcomes included maximal circulating tumor cells value, positivity (cutoff: at least 1 and at least 5 tumor cells/7.5 ml blood), and the association between natural killer cell activity and tumor cell counts. This trial was registered with (NCT02005770).

Results: Between March 2014 and April 2018, 210 participants were enrolled, assigned to sevoflurane (n = 107) or propofol (n = 103) anesthesia, and eventually included in the analysis. Anesthesia type did not affect circulating tumor cell counts over time (median circulating tumor cell count [interquartile range]; for propofol: 1 [0 to 4] at 0 h, 1 [0 to 2] at 48 h, and 0 [0 to 1] at 72 h; and for sevoflurane: 1 [0 to 4] at 0 h, 0 [0 to 2] at 48 h, and 1 [0 to 2] at 72 h; rate ratio, 1.27 [95% CI, 0.95 to 1.71]; P = 0.103) or positivity. In one secondary analysis, administrating sevoflurane led to a significant increase in maximal tumor cell counts postoperatively. There was no association between natural killer cell activity and circulating tumor cell counts.

Conclusions: In this randomized controlled trial investigating the effect of anesthesia on an independent prognostic factor for breast cancer, there was no difference between sevoflurane and propofol with respect to circulating tumor cell counts over time.


Breast cancer represents a major health issue: with more than 2 million new cases worldwide,[1] it is the most frequently diagnosed tumor and the leading cause of cancer deaths in women.[2] Despite primary treatment, between 6% of patients with localized tumors and 22% with nodal extension will face recurrence at 5 yr.[3]

Most patients diagnosed with breast cancer undergo surgical treatment. There have been increasing concerns, however, that the perioperative period would promote tumor spreading, either directly (i.e., through tumor manipulation), or indirectly, because systemic inflammation may affect immune responses against tumor cells.[4] Evidence also suggests that anesthesia itself may contribute to distant spread: anesthetic drugs seem to interfere directly with tumor cell biology and to decrease natural killer cells cytotoxic activity, which plays a critical role in tumor cell destruction and tumor growth.[5,6]

Although these effects have been well documented in preclinical studies, their relevance in the clinical setting is still matter of debate: intravenous anesthesia has been suggested to result in better survival rates compared with inhalational anesthesia, but evidence was mostly driven by retrospective analyses, which are prone to important methodologic limitations.[7–14] Conflicting findings also emerged from a few randomized controlled trials suggesting no effect on survival, but sample sizes were small, follow-up duration was short, and multiple interventions were evaluated without an adequate control group.[15–17]

Large, well designed randomized controlled trial are thus needed to clarify the effect of anesthetic drugs on cancer prognosis, but long follow-up periods often undermine the feasibility of such studies. To overcome this issue, the use of biologic markers as surrogates for prognosis may represent a valuable approach.[18] Among others, the presence of circulating tumor cells in the peripheral blood has been identified as a particularly promising indicator.[19] Hematogenous dissemination seems to occur long before clinical or radiological signs of metastases develop,[20] which places circulating tumor cells at an ideal location in the causal pathway leading to distant disease.[21] There is also increasing evidence that circulating tumor cells are independently associated with a higher risk of disease recurrence and with reduced survival, both in nonmetastatic and metastatic breast cancer.[22,23] In this respect, circulating tumor cell monitoring may represent a promising approach to better understand the effect of anesthesia on tumor behavior during the perioperative period.

Therefore, we conducted a randomized controlled trial to evaluate the effect of intravenous (i.e., propofol) versus inhalational (i.e., sevoflurane) anesthesia on postoperative circulating tumor cell counts in primary breast cancer patients. A superiority design was used to test the hypothesis that postoperative circulating tumor cell counts would be higher in patients receiving sevoflurane. The association between immune cell responses (i.e., natural killer cell cytotoxic activity) and circulating tumor cell counts was assessed in an exploratory in vitro study nested within this trial.