RF ablation with adjuvant therapy: Comparison of external beam radiation and liposomal doxorubicin on ablation efficacy in an animal tumor model

Abstract

Purpose: To determine the critical thermal dosimetry and relative efficacy for RF ablation combined with external beam radiation (XRT) or liposomal doxorubicin (LD), in an animal tumor model.

Materials and methods: This study was performed in two phases, in 13–18 mm diameter R3230 tumors subcutaneously implanted into Fischer rats. In phase 1, tumors (n = 30) were randomized into six groups. RF energy (titrated to 70°C tip temperature) was applied for either 2.5 or 5 min (n = 15, each group). For each duration, one of three adjuvant therapies was applied (n = 5, each): no therapy (control), LD (1 mg intravenously, 30 min post-RF), or XRT (20 Gy at 1 Gy min⁻¹, within 2 h post-RF), with sacrifice at 48 h for pathologic analysis. In phase 2, thermal mapping was performed in 20 tumors throughout RF application (70°C; 5 min), at 1.5–7 mm distances from the active electrode tip. Temperature profiles throughout the tumor were constructed and were used to interpolate temperatures over time at the critical ablation margin, to derive maximum threshold temperature, AUC (area under the curve) and CEM₄₃ (cumulative equivalent minutes at 43°C). Ablation sizes and all calculated values were compared within and across experimental groups using MANOVA statistics with pair-wise T-test for individual comparisons.

Results: RF/XRT produced the largest coagulation (11.7 ± 1.5 mm at 2.5 min, ≥15 ± 0.7 mm at 5 min), followed by RF/LD, and then RF alone (p < 0.001 for all comparisons). RF/XRT demonstrated temperature threshold decreases from RF alone of 11.7 ± 0.01°C and 12.7 ± 0.38°C at 2.5 and 5 min respectively (with absolute thresholds of 42°C for XRT compared to 52°C for RF alone). RF/LD had decreases of 4.0°C at 2.5 min and 4.4°C at 5 min. Thermal dose requirements (AUC) decreased by 7.79% or 9.28% for RF/LD compared to ≥19.36% or 25.82% for RF/XRT at 2.5 and 5 min (p < 0.001). CEM₄₃ values followed similar patterns (p < 0.001), but with a reduction of 10¹ and 10⁴ in magnitude for RF/LD and RF/XRT therapies at 5 min, respectively.

Conclusions: For a standardized RF dose, the combination of high dose XRT and RF increased ablation size compared to RF and liposomal doxorubicin or RF alone. Increased ablation size is more closely associated with decreased temperature threshold necessary to induce coagulation, rather than the total thermal dose.

• Radiofrequency
• liposomal doxorubicin
• external beam radiation
• tumor ablation
• hyperthermia

Introduction

Radiofrequency (RF) ablation is a rapidly maturing minimally invasive image-guided tumor therapy that is increasingly utilized for the treatment of focal hepatic, renal, bone and lung tumors [1–6]. However, RF treatment efficacy is hindered by its inability to effectively treat tumors greater than 3–5 cm in diameter [7–9]. This limitation has led to the development of combined therapies that aim to increase coagulation by modulating biophysical properties such as tumor electrical and thermal conductivity as well as blood flow [10–16], or modulation of thermal sensitivity to heat with adjuvant therapies [17].

Along these lines, several animal studies combining RF ablation with liposomal doxorubicin demonstrate increases in coagulation and reduced tumor growth rates over either therapy alone [18],[19]. Studies further suggest a synergy between these combined therapies and a lower thermal dose requirement to achieve coagulation [20]. However, the precise mechanism remains unknown. Additionally, clinical treatments combining RF ablation and other thermal ablation therapies including either hepatic artery infusion chemotherapy with floxuridine and 5-fluorouracil or chemoembolization with doxorubicin have produced improved local control with tolerable side effects [21].

More recently, the effects of combined RF ablation and radiation therapy have been explored. In early clinical studies, Grieco et al. demonstrated the effects of combined external beam radiation (XRT) and RF treatment in primary non-operable non-small cell lung cancer [22]. Similarly, Horkan et al. [23] established RF and radiation therapy dose-dependent tissue sensitivity associated with the combination therapy in a small animal tumor model, though the effects of combined therapy, particularly thermal dosimetry, must be further defined. Horkan et al. further suggested that combination RF/radiation therapy had greater endpoint survival and local cure rates, compared to RF/liposomal doxorubicin or RF alone [23]. Specifically, while D’Ippolito et al. [24] observed a 20% local tumor control rate with RF/liposomal doxorubicin, while, Horkan et al. [23] demonstrated an 82% local cure rate and 120 day survival rate of 50% when combining RF ablation instead with high dose (20Gy) external beam radiation in a similar animal model. However, to date, no head-to-head comparison between these therapies has been made.

The purpose of this study, therefore, was to determine the critical thermal dosimetry for tumor ablation combined with external beam radiation or liposomal doxorubicin, while simultaneously determining the relative efficacy of RF combined with external beam radiation versus RF ablation combined with an optimized liposomal doxorubicin regimen in a controlled animal model.

Materials and methods

Overall experimental design

The study was performed in two phases. In phase 1, animals (n = 30) were randomized into six treatment groups to determine the amount of coagulation necrosis that was induced by different combined treatment regimens. Tumors measuring 13–18 mm received RF energy (e.g. electrode tip temperatures titrated to 70°C) for 2.5 min or 5 min (n = 15, each group). At each RF duration, tissues received one of three potential additional adjuvant therapies post–RF: no adjuvant therapy (control) (n = 5), RF combined with liposomal doxorubicin (LD) (n = 5), and RF combined with radiation therapy (n = 5). Liposomal doxorubicin was administered in a 1 mg dose via IV, 30 min post-RF ablation. XRT was applied at a 20 Gy dose at a rate of 1 Gy per minute, within two hours post-RF ablation and animals were sacrificed 48 hr after treatment for pathologic studies. In the second phase of the study, a thermal mapping study was also performed in a second set of animals (n = 20) to determine heating profiles over time throughout tumors receiving the defined course of RF ablation, as described below. Together, these experiments enable coagulation diameter, maximum temperature, and two methods of calculation of total thermal dose at the margin of ablation (AUC and CEM₄₃ [25–27]), to serve as endpoints for comparison.

Animal model

The Institutional Animal Care and Use Committee approved this protocol. For all experiments and procedures, anesthesia was induced using intraperitoneal (IP) injection of a mixture of ketamine (50 mg/kg; Ketaject; Phoenix Pharmaceutical, Inc., St Joseph, MO) and xylazine (5 mg/kg; Bayer, Shawnee Mission, KS). When necessary, booster anesthesia injections at one-tenth the dose were administered every 30–60 min IP. The experiments were performed with a well established tumor model [17],[28], the R3230 mammary adenocarcinoma in Fischer female rats, the strain in which the tumor was originally derived (mean weight, 150 ± 20g; age range 7–9 weeks) (Taconic Farms, Germantown, NY). All steps in tumor implantation in the animal model have been previously described in several studies using this model [18–19].

Radiofrequency application

A 500 kHz RF generator (CC-1, Radionics®, Burlington, MA) was used to apply conventional, monopolar RF (Figure 1) using a 1 cm tip, 21 gauge electrode placed in the center of the tumor, as has been previously described [17],[18]. Each animal was placed on a standardized grounding pad (Radionics®), to complete the RF circuit, RF energy was applied for 2.5 or 5 min as required with the generator output titrated to maintain a designated tip temperature (70 ± 2°C, mean 90.4 mA ± 25.8, range 48–160 mA).

Figure 1. Rat tumor model. This figure demonstrates the experimental set up of RF ablation within the R3230 breast tumor model where (A) illustrates the grounding pad, (B) the RF generator, and (C) denotes the tumor with RF needle inserted. Inset: This illustrates the RF electrode A and thermocouple B inserted into an acrylic marker for insertion at specific radii to enable thermal mapping.

Liposomal doxorubicin ddministration

PEG stabilized, long circulating, liposomal doxorubicin (Doxil®; ALZA Pharmaceuticals, Palo Alto, CA) was administered via direct femoral injection using a dosage of approximately 8 mg/kg (1 mg in 500 µl dilution). The dosage was chosen based upon previous studies demonstrating a marked coagulation increase to 13.5 mm following RF ablation using prior parameters [18]. Injection occurred 20–30 min post-RF application, based upon time course studies performed by Goldberg et al. demonstrating maximal synergistic effect with this protocol [17].

Radiation therapy

External beam radiation was performed at a dosage of 1.0 ± 0.1 Gy for a duration of 20 min, giving a total 20 Gy dose over time, based upon prior work by Horkan et al. [23] using the same tumor model. Animals were positioned on their sides on a lead sheet, with a slit to allow radiation delivery. Subsequently, the sheet was inverted and a second lead sheet was added for lung protection. An RT 250 irradiator (Philips, Eindhoven, Netherlands) was operated at 250 kV and 12 Amps, using a filtration of 0.4 mm tin and 0.25 mm copper. The distance from the radiation source to the skin was 32 mm.

Thermal mapping studies

Thermal mapping was performed in 20 separate animals with 13–18 mm tumors to determine the heating history throughout the entire tumor during ablation, as a means of determining the critical ablation parameters of minimum threshold temperature and thermal dose. Animals were divided into five groups (n = 4) where RF ablation was performed at 70°C tip temperature for duration of 5 min. Thermocouples measuring temperature (°C) were placed at five different radii from the active electrode tip: 1.5, 2.5, 3.5, 5.0, and 7.0 mm (Figure 1 inset). Temperatures were recorded in intervals of 30 sec throughout the duration of ablation. Temperatures at each time interval and distance were averaged and maximum temperatures determined. For every 30 sec, negative exponential curves were calculated. Temperatures at the critical ablation margin, the interface between white coagulation and unablated tumor tissue for all 30 specimens of Phase I, were interpolated from these curves and the total heating history was constructed. Given the radial heating of RF ablation, this is the margin where one sees ablative effects with the lowest thermal dose [20]. This mapping study was used to derive three key endpoints: maximum threshold temperature (the isotherm at which coagulation was seen at the margin of coagulation at the conclusion of the ablation); AUC (area under the curve), which represents total thermal energy input (i.e. an integration of temperature over time) at the margin, and CEM₄₃ (cumulative equivalent minutes at 43°C) using the formula:t = time interval, T = average temperature during time interval t, R = number of minutes needed to compensate for a 1° temperature change either above or below the break point.

Quantification of tumor coagulation necrosis and histologic studies

Induced coagulation necrosis (tumor destruction) was measured 48 h after treatment, based upon results from prior studies [29],[30]. Animals were sacrificed with pentobarbital overdose, and tumors were excised and sectioned. In addition to hemotoxylin-eosin staining, histopathologic studies included staining for mitochondrial enzyme activity by incubating representative tissue sections for 30 min in 2% 2,3,5-triphenyl tetrazolium chloride (TTC) at room temperature. This latter test was used to identify irreversible non-specific cellular injury [31] during the early stages of RF-induced tissue necrosis [28],[32], and has been previously shown to correlate with histopathologic findings of RF-induced tumor coagulation [28],[30]. Gross measurements of tumor destruction were performed on both TTC-stained and unstained sections, and the extent of visible coagulation was measured with calipers (Figure 2). Coagulation diameter (longest measurement perpendicular to the inserted electrode [28] was determined by consensus of two observers who jointly measured and verified the findings. Previous studies have documented close correlation between gross pathologic and histopathologic findings for RF-induced coagulation [33].

Figure 2. Cross section of stained ablated tumor. This figure demonstrates a cross section of RF ablated tumor stained with TTC. Sharp demarcation is noted at the margin (arrow) between the white, ablated tissue and the unablated surrounding tumor tissue stained pink. Central dot in the white coagulation denotes the original position of the RF electrode.

Statistical and data analysis

For each experimental group, coagulation diameters were measured at 48 h and correlated to calculated values. Results were compared using MANOVA statistics with pair-wise T-test for individual comparison. Calculated threshold temperatures corresponding to the AUC (i.e. the total heating history of the tissue (°C·sec)) were also compared among the groups using MANOVA.

Results

Effects of combined therapies on RF coagulation

Significant differences in coagulation diameter were identified among the groups as compared with controls (p < 0.001) (Table I). RF combined with radiation therapy produced the greatest amount of tumor destruction (11.7 ± 1.5 mm at 2.5 min, ≥15 ± 0.7 mm at 5 min, (i.e. ablation consistently extending to the tumor margins)). Comparatively, RF combined with liposomal doxorubicin produced less coagulation (8.6 ± 0.2 at 5 min) than radiation therapy, yet more than the control treatment (6.7 ± 0.2 at 5 min; p < 0.001) (Table I). Differences in duration of treatment also produced differences in coagulation diameter. For example, a difference of at least 3 mm was exhibited between 2.5 min (11.7 ± 1.5 mm) and 5 min (≥15 mm) for combined treatment of RF and radiation therapy.

Table I.  The effects of application duration upon combined adjuvant therapies. This table demonstrates that an increase in RF dose produces increases in coagulation, AUC, and CEM₄₃. In comparison to RF alone and RF combined with Doxil, high dose XRT produces increases in coagulation while decreasing AUC and CEM₄₃ (percentage decreases in AUC are based upon the baseline of RF alone for a given duration). Within each experimental group, duration of application had no effect on temperature threshold for ablation.

Treatments		Coagulation diameter (mm)	Temperature threshold (°C)	Thermal dose (AUC) (°C/s)	Decrease in AUC (%)	CEM43	Decrease in CEM43 (%)
2.5 minute RF	RF control	5.3 ± 0.1	52 ± 0.40	6862.9 ± 36.08		978.5 ± 207.4
	Doxil	7.4 ± 0.2	47.2 ± 0.39	6226.3 ± 49.43	9.28	38.5 ± 5.57	96.07
	XRT	11.7 ± 1.5	40.3 ± 0.39	5090.7 ± 25.8	25.82	0.3 ± 0.04	99.99
5 minute RF	RF control	6.7 ± 0.2	52.4 ± 0.38	13987.5 ± 138.07		2124.1 ± 245.69
	Doxil	8.6 ± 0.2	48 ± 0.30	12898.2 ± 110.94	7.79	103.2 ± 34.76	95.14
	XRT	≥15 ± 0.7*	≤39.7 ± 0.35	∼11167 ± 161.9	≥19.36	0.46 ± 0.05	99.98

*Ablation to the tumor margin for 5 min RF with radiation.

Thermal dosimetry

Classic sigmoidal Boltzmann curves were obtained for each temperature time curve at distances of 1.5–7 mm (r² = 0.94–0.99) (Figure 3) [20]. Calculations based upon these temperature measurements demonstrated significant decreases in the ablation temperature threshold (p < 0.001) for combined therapies. The temperature required for ablation decreased by 11.7 ± 0.01°C (from 52°C to 40.3°C) and 12.7 ± 0.38°C (from 52.4°C to 39.7°C) for RF application combined with radiation therapy for 2.5 and 5 min, respectively. By comparison, RF/liposomal doxorubicin, yielded only a 4°C decrease (i.e. a 47.2°C threshold) at 2.5 min and 4.4°C decrease (i.e. a 48°C threshold) at 5 min.

Figure 3. Thermal mapping of temperature distributions within a subcutaneous tumor. Mean temperature profiles for multiple measurements at five distances within the tumor (1.5–7 mm from the electrode) are presented with T-bars. The temperatures increase in a sigmoidal Boltzmann distribution (R2 = 0.94–0.99).

Correlation of thermal dose (AUC and CEM₄₃)to coagulation

A reduced thermal dose requirement for RF/liposomal doxorubicin and even greater reduction for RF/radiation therapy was observed at both durations of RF application compared to RF alone (p < 0.001). However, all therapies performed at durations of 5 minutes yielded greater AUC and CEM₄₃ values than those performed at 2.5 min (p < 0.001). The percentage decrease in AUC values for each combined therapy were similar, regardless of duration, in comparison the RF control. AUC values for RF applications decreased by 7.79% and 9.28% when followed by liposomal doxorubicin injection, and by at least 19.36% and 25.82% when followed by radiation therapy at 2.5 and 5 min, respectively (Table I). CEM₄₃ values followed similar patterns (p < 0.001), but with a reduction of 10¹ and 10⁴ in magnitude for liposomal doxorubicin and radiation therapies at 5 min, respectively.

Discussion

Classically, studies combining external beam radiation therapy and/or chemotherapy with low temperature hyperthermia (treatment duration of 1–2 hours at 42–45°C [27],[34]) have suggested synergy and increased efficacy for treatment of tumors in the lung [35],[36], prostate [37], bladder [38],[39], and other sites [40]. In contrast, targeted minimally-invasive tumor therapy with RF ablation uses focal heating of the tumor to extremely high temperatures, though large areas of the tissue beyond the zone of coagulation are still exposed to non-coagulative hyperthermic temperatures (40–50°C) [20]. Therefore, more recently, several preliminary studies have combined RF ablation with adjuvant therapies such as external beam radiation and IV chemotherapy to increase tumor destruction by modulated thermal sensitivity of tumor cells in these peripheral zones exposed to non-lethal temperatures [19],[22],[23],[30]. Given the increased clinical adoption of many of these combinations, continued investigation into both the specific mechanisms of combined therapies and clinical optimization strategies, are required. Along these lines, our study further defines the relationship between radiation therapy and RF ablation, as well as its greater efficacy in comparison to RF combined with doxorubicin.

This six-arm study varying RF dose and combined adjuvant demonstrated the varying efficacy of radiation therapy and liposomal doxorubicin in this tumor model. As has been previously demonstrated in this model, combination therapies result in greater coagulation compared to RF alone. In this study, we further observed significantly greater coagulation for treatment of RF/radiation therapy compared to RF/liposomal doxorubicin, and with both combination therapies over RF alone. These findings, especially the suggestion of complete tumor coagulation observed with RF/radiation therapy at 5 min, are consistent with a prior study by Horkan et al. in which most tumors had complete local tumor regression [23].

Significant differences in the temperature threshold to achieve ablation were also observed between various therapies at both durations, and were ∼40°C for RF/radiation therapy,∼48°C for RF/liposomal doxorubicin, compared to ∼52°C for RF alone. These findings suggest that increased coagulation with RF/radiation therapy occurs due to the ability of adjuvant radiation therapy to induce greater damage to tumor cells in the peripheral zone exposed to previously reversible injury from non-lethal temperatures [34]. While the exact underlying mechanisms require further investigation, the difference in temperature threshold between adjuvant radiation and liposomal doxorubicin likely reflects differences in mechanism of cell injury, or variable effects on an as yet ill-defined mechanism, with radiation combined therapy being a greater potentiator for cell death at lower thermal dose. Indeed, further investigation into possible underlying mechanisms such as the increased free radical and/or heat shock protein formation that leads to greater tumor necrosis, the role of oxidative stress [41], and changes in programmed cell death (apoptosis) [42], is required. Differences in the uniformity of tissue effects, given that interstitial distribution of chemotherapy is likely more heterogeneous in distribution compared to the effects of radiation therapy, is also a likely contributory factor.

AUC, measuring the total heating curve at the margin of coagulation, a conventional measure of total thermal dose, had a similar percentage decrease with either combination radiation (19.36–25.82%) or liposomal doxorubicin (7.79–9.28%) for both RF durations. A similar pattern was observed for CEM₄₃, though this measurement is likely to have even more limited clinical application for ablation given the very wide variability in magnitude (up to 10⁴) in this study, similar to widely variable results (10¹¹) reported by Mertyna et al. in various ex vivo and in vivo tissues [20]. Yet our results also suggest that classic hyperthermia rules may not apply to higher temperatures and shorter durations of RF combined treatment given the fact that thermal dosimetry changed with time, whereas it has been reported as being constant for classic low temperature hyperthermia scenarios. As a result, temperature endpoints should likely be relied upon in the clinic at this time, realizing that wide variations in thermal sensitivity have been reported for different tissues [20].

Some limitations in our study warrant noting. Our results were obtained through experimentation using the R3230 breast tumor model, thus our conclusions are primarily applicable to this tumor type. Thus, it is conceivable that a combination of RF and chemotherapy may be more beneficial than RF with radiation therapy for other clinical and relevant tumor types such as hepatocellular and renal cell carcinoma, which may be heterogeneous (i.e. cystic) or hypervascular [43],[44]. Furthermore, we only included a single chemotherapy agent, liposomal doxorubicin, in this study, and presumably this tumor line may demonstrate different sensitivities to other agents. Indeed, a prior study combining RF ablation with different liposomal chemotherapeutic agents in this same animal model has demonstrated varying tumor sensitivity to various agents [19]. Along similar lines, we use a single radiation dose, and while this has been previously identified as the optimal dose [23], differing and more clinically realistic radiation doses may not be as effective compared to chemotherapy. Additionally, interpolative thermometry in this study was facilitated by the orderly decrease in heat distribution observed in this relatively homogeneous tumor model. Yet, prior studies have suggested that variable tumor composition or tumors with high perfusion demonstrate non-uniform heating surrounding RF electrodes, with documented rapid exponential decreases in tissue temperature [10],[45]). Furthermore, major heat-sink effect from blood vessels will decrease the level of hyperthermic sensitization of tumor to irradiation [46],[47]. Thus, further work and tailoring is likely warranted.

In conclusion, this study demonstrates that the combined therapy of high dose radiation therapy and RF markedly increases ablation size in comparison to traditional RF or RF combined with liposomal doxorubicin. Additionally, the correlation between ablation size and specific threshold temperature at the ablation margin suggests that temperature, rather than thermal dose, may serve as a more applicable marker for successful RF application in a clinical setting. However, this study further supports the strong clinical potential clinical of combined adjuvant therapies, necessitating further research into mechanisms of operation.

Acknowledgements

Supported by grant number NCI 1R01CA112533–01 “Enhanced RF Tumor Ablation with Liposomal Chemotherapy” from the National Institutes of Health, USA.