An In Vivo Mouse Model of Total Intravenous Anesthesia During Cancer Resection Surgery

More than 60% of patients with cancer receive anesthesia for surgical resection¹. Currently, there are no specific clinical guidelines that determine the choice of anesthesia used in cancer patients. Surveys of anesthesiologists indicate a preference for volatile-based anesthesia, including during cancer surgery^2,3. However, there is a growing body of evidence that the use of propofol-based total intravenous anesthesia (TIVA) during cancer surgery may associate with improved postoperative outcomes (progression-free survival, overall survival) when compared to volatile anesthesia⁴. Subsequent clinical studies continue to report contradicting results^5,6,7,8. These findings support the need for preclinical studies to better understand the mechanistic effects of different anesthetic agents on cancer-related outcomes.

However, in in vivo studies that model cancer surgery, anesthesia is frequently an incidental part of the procedure. The rationale for the choice of anesthesia is often not the focus of the experimental design, and its impact on cancer-related endpoints may not be evaluated. For example, in vivo studies that require maintenance of anesthesia for cancer surgery most commonly use inhaled volatile anesthesia⁹. Where propofol has been used in in vivo studies, it has been delivered by single bolus dosing with intraperitoneal delivery, which does not replicate clinical onco-anesthetic protocols¹⁰. That approach of propofol administration induces light anesthesia that is suitable for rapid procedures. However, it does not allow maintenance of anesthesia that is required for cancer resection surgery which may be protracted. Furthermore, the absorption kinetics of intraperitoneal delivery is distinct to clinical methods of administration.

A model of propofol-based TIVA for cancer resection surgery was developed to address this need. A protocol for sustained maintenance of anesthesia with titration of the anesthetic agent to allow response to the surgical stimulus was developed to replicate key aspects of anesthetic delivery to patients having cancer surgery. The resulting protocol is used with a mouse model of cancer to provide TIVA during cancer resection surgery. The effect on short-term and long-term cancer-related outcomes is evaluated.

All animal studies were undertaken under the approval of the Institutional Animal Care and Use Committee at Monash University. In this study, female Balb/c mice aged 6-8 weeks were used.

1. Prepare cancer cells

1. Culture tumor cells in medium. Culture 66cl4 murine mammary cancer cells in alpha-MEM containing 10% FBS and 200 mM glutamine. Cells should text negative for mycoplasma. OPTIONAL: Use cells that are stably transduced to express firefly luciferase for bioluminescence imaging to enable cancer recurrence monitoring after resection surgery¹¹ (see step 4).
   NOTE: The cell line and the medium mentioned above were used in this study.
2. Grow cells at 37 °C with 5% CO₂. Passage cells aseptically in a hood at <80% confluency. Use low-passage cells in the logarithmic growth phase for optimal in vivo results.
3. Lift the adherent cells with 0.5 mg/mL of trypsin in PBS with 10 mM EDTA; 2 mL for a T75 flask. When lifted, add medium to inactivate trypsin, and wash the cells in PBS.
4. Count the cells using a hemocytometer. Dilute the cells in PBS for injection. For 66cl4 mammary cancer cells, inject 1 x 10⁵ cells in 20 μL of PBS per mouse.
5. Place the cells on ice prior to the injection.

2. Generate a mouse model of breast cancer

1. Use 4% isoflurane to anesthetize the mouse in an induction chamber. Then, maintain anesthesia with 2%-3% isoflurane using a nose cone. Confirm proper anesthetization by lack of response to toe pinch.
2. Prepare the injection site by wiping the fourth left mammary fatpad area using a single-use alcohol swab.
3. Draw up tumor cells (see step 1.4) into a 25 μL Hamilton syringe attached to a sterile 27 G hypodermic needle.
4. Inject the cells into the fourth left mammary fatpad. Use forceps to secure and lift the skin. Inject approximately 1 mm from the nipple.
5. OPTIONAL: If cells are tagged with luciferase, confirm successful injection of tumor cells by bioluminescence imaging. Inject 100 μL of 150 mg/kg D-luciferin into the lateral tail vein of the anesthetized mice using a 0.5 mL insulin syringe with a 30 G hypodermic needle.
6. OPTIONAL: Place the mouse in a bioluminescence imaging system with the mammary fatpad facing up. Wait for 2 min from the luciferin injection for optimal tissue uptake of luciferin, then image for 10 s.
7. Place the mouse in a clean cage and allow it to recover from anesthesia.
8. Continue to monitor animal welfare as per the institutional animal ethics guidelines.

3. Induce stable anesthesia with intravenous delivery of propofol

1. Monitor growth of the primary tumor using caliper measurement and calculate the tumor volume using the equation: Volume (mm³) = (length x (width)² ÷ 2).
2. Perform tumor resection surgery on mice when the primary tumor reaches the required volume (here, 80-90 mm³).
3. Set up an automated syringe pump with a 30 G 1 mL insulin syringe containing propofol formulation (2% Lipuro propofol) (Figure 1A).
4. Induce anesthesia of the mouse in an induction chamber with 3% sevoflurane or isoflurane.
   NOTE: Here, sevoflurane was used as this is the predominant volatile used clinically.
5. Transfer the mouse to a 37 °C heating pad for the duration of the surgery. Briefly maintain anesthesia with 2%-3% sevoflurane using a nose cone.
6. Apply aqueous lubricant to the eyes to prevent drying. 
7. Inject 0.05 mg/kg of buprenorphine subcutaneously for analgesia.
8. To prepare for surgery, shave the abdomen and prepare the skin for surgery with an iodine-povidone solution. Wipe the skin using an alcohol or sterile wipe.
9. To deliver propofol-based TIVA, cannulate the lateral tail vein using a sterile 30 G hypodermic needle attached to a sterile polyurethane catheter. Confirm correct placement by blood flashback into the catheter (Figure 1B). Adjust the delivery of sevoflurane as required during intravenous cannulation to maintain a stable depth of anesthesia demonstrated by loss of corneal and pedal reflex, and respiratory rate < 100 breaths per min.
10. Commence propofol-TIVA by administering 2% propofol as an initial bolus of 27 mg/kg for over 1 min. Cease sevoflurane administration.
11. Continue the infusion of propofol at a maintenance rate of 2.2-4.0 mg/kg/min to maintain a stable depth of anesthesia for the duration of the surgery (Figure 1C).

4. Resect the primary tumor

1. Make a straight 1 cm incision inferior to the tumor in the region of the left fourth mammary fatpad. Carefully resect the tumor and the left inguinal lymph node using dissection with blunt forceps.
2. OPTIONAL: If using luciferase-tagged tumor cells, use bioluminescence imaging to confirm clear surgical margins. Inject 150 mg/kg D-luciferin into the lateral tail vein, wait for 2 min, and then image for 60 s using a bioluminescence imaging system. If a residual tumor is identified, resect additional tissue from the mammary fatpad and re-image to achieve clear margins.
3. Ensure hemostasis at the surgical site and close the skin using 5-0 nylon sutures. Ensure that the fur does not enter the surgical site.
4. At the conclusion of resection surgery, cease anesthesia. Place the mouse in a clean cage on a 37 °C heating pad and allow it to recover from anesthesia.
5. Monitor every 15 min after anesthesia until the mouse has returned to normal alertness. Then, monitor the mouse every 12 h for 48 h after surgery.
6. Administer 0.05 mg/kg of buprenorphine subcutaneously every 12 h for 48 h after surgery.
7. After 7-10 days, remove sutures using sterile curved stitch cutters under brief sevoflurane or isoflurane anesthesia.

5. Track cancer recurrence with in vivo imaging

1. Use bioluminescence imaging to track cancer recurrence after resection surgery non-invasively. Use a bioluminescence imaging system to monitor the mice once per week for evidence of primary tumor recurrence or distant recurrence, commencing the week following surgery.
2. Induce anesthesia of the mouse in an induction chamber with 4% isoflurane. Then, transfer the mouse to a 37 °C heating pad for the duration of anesthesia and maintain anesthesia with 2%-4% isoflurane using a nose cone.
3. Apply aqueous lubricant to the eyes to prevent drying.
4. Inject D-luciferin 150 mg/kg into the lateral tail vein. Wait for 2 min, then measure bioluminescence over a 60 s exposure to detect recurrence of the primary tumor or distant metastasis.
5. If the primary tumor recurs and becomes palpable, commence monitoring of tumor growth using caliper measurements.
6. At the end of the experiment, humanely kill the mice according to the approved protocol. Here, CO₂ was used, followed by cervical dislocation.

This method describes a model of total intravenous anesthesia (TIVA) with propofol during cancer resection surgery in mice. Propofol is delivered in this mouse model through an intravenous catheter using a syringe pump (Figure 1A,B) to replicate delivery of TIVA in the clinical setting of anesthesia for cancer surgery. Use of the syringe pump minimizes exposure to volatile anesthesia by allowing rapid conversion from initial induction by inhalational anesthesia to intravenous delivery.

After stable anesthesia was achieved by propofol-based TIVA, the primary mammary tumor was resected. In vivo bioluminescence imaging was used to confirm complete resection of the primary tumor (Figure 2). Regular monitoring of mice by non-invasive in vivo bioluminescence imaging identified distant recurrence by luciferase-tagged tumor cells in the lung (Figure 2). This method is also suitable to track local recurrence in the mammary fatpad.

In addition to tracking long-term events such as recurrence, the model may be used to assess events that occur during the perioperative period. These early events may provide mechanistic insight into the effects of anesthesia and other surgical factors on cancer-related outcomes. Twenty-four hours after the cancer surgery under propofol, a multiplex enzyme-linked immunosorbent assay was used to quantify circulating plasma cytokines (Figure 3). Cytokines were evaluated in 7 mice; the appropriate group size will be influenced by the effect size of the endpoint of interest.

Figure 1: Experimental set-up for propofol-based TIVA. (A) A syringe pump is used to ensure the controlled delivery of propofol from a 1 mL insulin syringe. (B) Propofol is delivered into the lateral tail vein by intravenous catheter, connected to a syringe pump, via a 30 G needle. The asterisk shows the needle insertion point. (C) Schematic illustrating the plasma concentration of propofol achieved by sequential administration of a bolus followed by a constant infusion using the syringe pump, compared to delivery by bolus or infusion only. Please click here to view a larger version of this figure.

Figure 2: Cancer progression after surgical resection of the primary mammary tumor under propofol-based TIVA. Non-invasive bioluminescence imaging of luciferase-tagged tumor cells was used to track initial growth of the primary tumor growth, successful surgical resection, and subsequent distant recurrence to the lung. Please click here to view a larger version of this figure.

Figure 3: Circulating cytokine levels measured after cancer resection under propofol-based TIVA. Multiplex enzyme-linked immunosorbent assay was used to quantify plasma cytokines 24 h after surgery. Each data point represents data from one mouse. Lines show mean and standard error (N = 4-7). Please click here to view a larger version of this figure.

This study reports on a protocol for administering total intravenous anesthesia (TIVA) with propofol in a mouse model of breast cancer that replicates key aspects of clinical practice for TIVA in patients requiring cancer surgery. The protocol allows investigation of both short-term and long-term clinically relevant outcomes after cancer surgery in a mouse model of cancer progression, including measurement of cytokine levels and cancer recurrence (Figure 2, and Figure 3). The methodology will be useful for the evaluation of the effects of TIVA on cancer-related outcomes and its comparison with other anesthetic techniques such as volatile anesthesia.

In contrast to existing protocols that deliver a single intraperitoneal bolus of propofol for sedation during minor interventions such as blood sampling delivery¹⁰, this protocol allows prolonged intravenous delivery of propofol for maintenance of anesthesia during major interventions such as surgery. Both the induction and maintenance dose of propofol was carefully titrated to optimize anesthetic depth suitable for major surgical interventions while minimizing mortality from hypotension or cardiac arrest. It was found that hypotension could be avoided by using a strict induction dose of 27 mg/kg administered over 60 s, with concurrent down-titration and cessation of inhaled sevoflurane. Maintenance was achieved using an infusion of propofol at 2.2-4.0 mg/kg/min. During major surgical interventions, such as cancer resection, titration within this range was important to respond to and obtund the magnitude of surgical stimulation. This replicates clinical practice and prevents either over-dosing anesthesia, which can result in hypotension or death, or under-dosing, which can result in emergence from anesthesia, movement, or surgical stress.

A limitation of the model is the brief use of volatile anesthesia to induce anesthesia prior to canulation. This approach was chosen because of the ease of cannulation after induction of anesthesia, allowing the rapid delivery of a therapeutic dose of propofol to continue anesthesia. In addition, mice were briefly anesthetized with volatile anesthesia for tumor cell inoculation, to remove suture, and for imaging. Sevoflurane was used for volatile anesthesia during resection surgery as it is often used in clinical practice. However, isoflurane is also used in clinical practice. Future studies may use a single agent for all episodes of inhalation anesthesia. With experience, less than 2 min elapsed from loss of righting reflex in response to volatile anesthesia exposure to commencement of propofol dosing. Nonetheless, for analyses that intend to compare propofol-based TIVA with inhalational volatile anesthesia techniques, interpretation may be confounded by even the brief use of volatile anesthesia.

An alternative approach to volatile anesthesia induction is to cannulate the lateral tail vein of an awake mouse. While not suitable for all situations, this may provide an alternative to induction by inhalation anesthesia. However, movement of the awake mouse may result in the cannula being dislodged, leading to failure of anesthesia induction. In addition, movement of the needle can result in extravasation of propofol from the vein, which puts the tail at risk of tissue necrosis. This has welfare implications for the mouse, and any associated adrenergic activation as a result of physiological stress may impact the validity of the observed results¹¹.

An additional potential limitation is the use of buprenorphrine as a postoperative analgesic. Opioids may modulate postoperative inflammatory and immune responses^12,13. Buprenorphine was chosen for analgesia as its effects on the immune response are less than for other opiates¹³. Nevertheless, future studies might consider the use of non-opioid analgesic agents.

Despite advances in oncological therapy, local and distant cancer recurrence can occur after surgery and remains a dominant cause of mortality in cancer patients. Many patients will be exposed to anesthesia, often multiple times, during diagnostic and therapeutic operative procedures. A growing body of evidence from in vivo and in vitro studies implicates anesthetic agents in modulating the perioperative response to surgery and impacting diverse aspects of tumor cell biology¹⁴. To better understand the impact of anesthetic agents on cancer progression, the model of intravenous propofol anesthesia developed here will be important in future mechanistic preclinical research. This model may be used to interrogate the mechanisms underlying the effects of anesthetic agents on immunomodulation, perioperative inflammatory response, and tumor cell growth and invasion. Furthermore, this model could be extrapolated for use in non-cancer surgery research where anesthetic agents may have effects on other systems, such as cardiac surgery, trauma research, or critical illness (e.g., sepsis) as propofol is a common sedation used in intensive care units.