Effects of energy-matched pulsed and continuous ultrasound on tumor growth in mice.

A diagnosis of cancer has been a contraindication to ultrasound (US) use,[1-3] but people with cancer may benefit from US if methods could be found that would allow for safe application of this modality. In a previous study, we investigated the effects on murine tumors of continuous therapeutic US delivered at an energy level of 5 W-min and an intensity sufficient to generate tissue temperatures of 39 [degrees] to 41 [degrees] C. We found higher final tumor volumes and weights in the group that received US as compared with a control group. The tumors exposed to US grew deeper, with larger volumes below the skin surface. We believe that this increased growth could be due directly to the energy delivered to the tissues by the ultrasonic wave and to the associated tissue temperature rise (TTR). Both mechanical and thermal effects of US could be contributory.

Using settings commonly used for obstetric US, Lejbkowicz et al[5] investigated the effects of US on normal and malignant cells in vitro. Cells were placed in insonation tubes, suspended in a thermostatically controlled water bath, and exposed to 2-MHz US at 0.33 W/[cm.sup.2] for 4 minutes. Findings from that study indicate that the growth of tumor cells was inhibited in the cells exposed to US. Lejbkowicz et al noted that cavitation, one of the mechanical effects of US, could explain the growth-inhibiting effects seen in their study. The growth inhibition of tumor cells seen in this in vitro study may not occur in vivo. Our previous in vivo study[4] showed an increase in tumor cell growth when the cells were exposed to higher energy levels of US. The US used in the study by Lejbkowicz et al was delivered at lower energy levels than those used therapeutically with pulsed or continuous US.

Pulsed US operates at the same carrier frequency as continuous US (usually 1 or 3 Mhz), but the signal is broken into "on" and "off" cycles similar to the duty cycles used with electrical muscle stimulators.[6] Because of these "off" cycles, pulsed US transmits less energy per unit of time and produces a reduced thermal effect, making it the appropriate choice when the nonthermal effects of US are desired. Kramer[7] showed that continuous US treatments were associated with an increase in tissue temperatures and that pulsed US treatments showed no tissue temperature elevation. He compared continuous US, pulsed US, and placebo US. He noted a similar pattern of increased subcutaneuos tissue temperature responses for the continuous US group and decreased tissue temperature responses for the pulsed and placebo US groups. Therefore, for the purposes of our study, the pulsed mode would be the appropriate choice for exposing murine tumors to US at lower tissue temperatures and energy levels.

The purpose of our study was to determine whether sonicated tumor growth is dependent on total infused US energy, and whether there is a difference if the energy is applied continuously or in a pulsed fashion. Specifically, we examined whether reducing US energy and tissue temperature by decreasing intensity or by using pulsed US would result in reduced tumor growth. We hypothesized that increases in tumor weight and volume would be greater in the US groups than in the control group.

METHOD

Pilot Study

A pilot Study using 17 female C57BL/6 mice was conducted using various dosages of continuous and pulsed US. Our goal was to determine whether the dosages of pulsed and continuous US selected for our study produced TTRs equivalent to therapeutic dosages for humans. The pilot study was performed similar to the pilot study described in our previous report.[4] The levels of US selected for this study from the results of our pilot study are listed in Table 1. Energy supplied to the tissue can be represented by the product of intensity X duration corrected for on/off cycle. The pulsed and low-power continuous US groups were matched to infuse equal amount of energy to the mice.

[TABULAR DATA 1 NOT REPRODUCIBLE IN ASCII]

Main Study

Female C57BL/6 mice (N=174), aged 6 to 8 weeks and weighing between 16 and 17 g each, were sedated with an intraperitoneal injection of 0.03-mL of Ketaset(*) (ketamine hydrochloride, 100 mg/mL) and Rompun([dagger]) (xylazine, 20 mg/mL), in a ratio of 10 mL of Ketaset to 1 mL of Rompun. The right flank of each mouse was shaved, and the right ear was tagged for identification. A single-cell suspension of methylcholanthrene-induced 205 rhabdomyosarcoma cells was prepared as previously described,[8] and 0.05 mL of a suspension of 5 x [10.sup.5] cells was injected subcutaneously into the animal's right flank.

After 2 weeks, when the single right-flank tumor reached 5 mm in diameter, each mouse was sedated and randomly assigned to one of three groups for the levels of US treatment shown in Table 1. The original numbers of mice in each group were as follows: 60 mice received low-power continuous US (LC US group), 63 mice received pulsed US (pulsed US group), and 51 mice were given no US (control group). Fourteen mice (6 mice in the LC US group, 3 mice in the pulsed US group, and 5 mice in the control group) did not complete the study because of early demise. Data for a fourth group of mice (n=37) that received high-power continuous US (HC US group) were collected in our previous study.[4] An Enraf-Nonius Sonopuls 434([double dagger]) was used to deliver the US through a 0.5-[cm.sup.2] treatment head placed directly over the tumor. The sound head's beam nonuniformity ratio maximum was 6.0. Enraf Sonogel([double dagger]) was used as a conduction medium. A device designed by the National Institutes of Health Biomedical Engineering and Instrumentation Program (BEIP) was used to position the animal on a platform under the US head. The US head remained stationary during treatment to provide consistent US application to all mice. The US machine was calibrated, tested, and found to yield reliable measurements by the BEIP.

Before initiation of US treatment, tumor dimensions were measured using a digital caliper (Sylvac Fowler Ultra-Cal II([sections])). One investigator (LSR), who was blinded to group assignment, measured all tumors and demonstrated acceptable intrarater reliability over three trials (intraclass correlation coefficient [ICC (1,1)]=.92). This ICC is based on the assumption that rater effect and rater X target interactions are inseparable from error.[9,10] Tumor dimensions were measured on the skin surface on days 1 (baseline), 7 (midtreatment), and 14 (preexcision). Based on our previous analysis of surface tumor' orientations, we used only two dimensions (ie,width and length) for measurements. Areas were calculated with the equation: A = ([Pi]/4) X ab, where "a" represents the width and "b" represents the length.[11] Areas were used to determine whether the tumors were growing in all groups during the study.

A thin layer of gel was applied to the tumor, the mouse was placed on the platform. and the US head was lowered to within 1 mm of the mass and held stationary. During the next 2 weeks, the experimental groups received a series of 10 3-MHz US treatments. Tumor tissue temperatures were not monitored during the study because we believed that the needle manipulation of the tumor could cause increased growth or metastasis. The control group received the same handling that the experimental groups received, including sedation and US gel application, but they did not receive US.

Prior to, during, and after treatment, the animals were housed in isolators, with a mix of control and experimental group mice in each cage. An air filter was in place on each isolator, and rodent chow and water were provided as needed. After the treatment period, all mice were killed with carbon dioxide gas.[12] Tumors were excised and weighed on an OHRW balance.([parallel]) They were again measured with calipers, and volume was derived from the three axial measurements of the tumors by using the ellipsoidal volume equation: V = ([Pi]/6) X abc, where a represents tumor width, b represents tumor height, and c represents tumor depth.[13] Tumors were fixed in 10% neutral buffered formalin. The right and left inguinal lymph nodes from each mouse were harvested, measured, and collected into formalin. All organs were examined for presence of metastases. All mice received complete gross necropsies by a veterinarian pathologist who was blinded to group assignment.

Data Analysis

The surface area measurements were assessed with a one-way repeated-measures analysis of variance (ANOVA), and the grouped data were used to confirm that the tumors grew continuously over the data collection period. A repeated-measures ANOVA was applied to the final weights and volumes of the excised tumors from the LC US, pulsed US, and control groups as well as those of the HC US group in our previous study.[4]

Results

Tumor Measurements

The projected areas of the tumors on the skin surface are presented in Table 2. There was an increase in the surface projections of the tumors for all groups combined over the period of data collection. The ANOVA results for weight and volume are shown in Tables 3 and 4. A Student-Newman-Keuls post hoc analysis revealed that the HC US group produced heavier (0.933 - 0.154 g [X [+ or -] SEM], P [is less than] .01) and larger (789 [+ or -] 140 [mm.sup.3] [X [+ or -] SEM]. P [is less than] .02) tumors than any of the other groups produced. There were no differences among any of the other groups. Final weights and volumes are summarized in Table 5 and illustrated in Figures 1 and 2.

[TABULAR DATA 2 & 4 NOT REPRODUCIBLE IN ASCII]

[FIGURES 1 & 2 GRAPHS OMITTED]

Table 3. Results of Analysis of Variance for Excised Tumor Weight

Source    df      SS      MS      F      P

Group      3     4.50    1.50   3.83   .01
Error     192   75.12     .39
Total     195   79.62

Table 5. Weights and Volumes of Excised Murine Tumors

                                                   Volume
                                   Weight (g)    ([mm.sup.3])
Group                              X     SEM     X     SEM

High-powered continuous US(a)    0.933   0.154   789   140
Low-powered continuous US        0.563   0.077   564    71
Pulsed US                        0.560   0.078   525    74
Control                          0.516   0.034   406    42

(a) US=ultrasound

Necropsy Results

There were no differences in tumor morphology among the groups. Histopathology of all groups was as noted in our previous study.[4] The tumors were predominantly noninvasive subcutaneous masses, with most tumors having small (1- to 3-mm-diameter) surface ulcers. Distant metastatic lesions were present in two of the HC US group animals, and no metastases were noted in any other group.

Discussion

Our study addressed the questions of whether tumor growth would be affected by the amount of US energy applied and whether there would be a difference if the energy were applied continuously or in a pulsed fashion. Data from our initial study involving the HC US group were included in this study for the purpose of comparison. The inclusion of these data is possible because both studies occurred within a year of' each other and the equipment, techniques, facilities, and researchers were identical. Both studies used C57BL/6 mice, which are laboratory bred to be as identical as possible.

Ultrasound dosage is related to total energy infused into the tissue. There are two ways to reduce US input. One way is to decrease intensity, and the other way is to shift from continuous input to pulsed input. Settings for treatment of the pulsed US group were selected to infuse the same amount of US energy that was administered to the LC US group. This selection was accomplished by increasing the intensity and duration of treatment when using the pulsed mode. Both increases were necessary. Treatment duration was limited to 15 minutes because of the limits on mouse sedation procedures, and intensity was kept below 2 W/[cm.sup.2] to remain within a realistic clinical range.

Tumor growth in the pulsed US group was the same as in the LC US group, but less than that found in the HC US group. The HC US group had greater energy infusion than the other groups had, and the mice in that group subsequently developed larger tumors. In a pilot study performed prior to our original study,[4] the HC US treatment produced higher TTRs than those produced in the other groups. The LC and pulsed US groups received identical energy infusion, but higher TTRs resulted for the group that received LC US in the pilot study performed prior to the present study. Because the LC and pulsed US groups did not produce different-size tumors, we cannot attribute the tumor growth to TTR alone. Both of these "low-energy" groups (LC US and pulsed US) demonstrated tumor growth more like the control group that the "high-energy" HC US group.

Our results suggest that tumor growth might be more dependent on accumulated US energy that on TTR and that pulsed US or low-intensity US might be less harmful as far as tumor involvement is concerned. Whether pulsed US is effective for treatment of the targeted conditions (presumably unrelated to the tumor presence) is beyond the scope of this study. The type of tumors used in our study, (ie, rhabdomyosarcomas) are known to metastasize easily. They are spread by lymphatic and hematogenous metastasis or by direct extension. A particularly high incidence of lymph node involvement appears to be associated with rhabdomyosarcomas of the extremities.[14] A greater incidence of metastasis in vivo might have been expected in our pulsed US group because pulsed US has been shown to dislodge cancer and cultured human cells in vitro.[15,16] This greater incidence of metastasis did not occur in our study because the only group showing any metastasis was the HC US group. Had the tumors been present for longer periods of time and received more than 10 US treatments, we may have seen more metastases.

A stationary sound head was used for consistency of application and because the small surface area of the tumor and mouse flank would not accommodate a moving sound head. Pilot study results showed that the use of the stationary sound head resulted in a mild, consistent temperature rise in the tumor tissues, similar to the increase noted in human tissues during therapeutic application. The US dosages used in our study were determined, through a pilot study measuring tumor tissue temperature to be equivalent to US dosages used clinically on human patients. Tumors in mice have been extensively studied, and although tumors in mice are similar to tumors in humans, they must still be acknowledged as being different. Therefore, the responses of murine tumors might not be identical to those of tumors in humans. Nothing in our study indicates that US causes tumors. Nevertheless, a difinitive study of ultrasonic effects on tumors using a human model does not appear to be possible because of ethical considerations.

Further research is needed in this area, and it would be interesting to study the response of other tumor types to US treatment. More research involving the effects of therapeutic US on malignant and normal cells is needed. This need for future research is supported by a recent study done by Forytkova et al[17] that examined the effects of US on DNA synthesis in murine tumor cells in vitro. Clinicians should note that Forytkova et al emphasized that the effects of US on DNA synthesis in normal cells remain unknown.

Clinical Considerations

Ultrasound is used as a treatment intervention for a variety of clinical conditions. In our Previous study,[4] we demonstrated that caution should be exercised when using US if cancers is suspected. In this study, we have shown that decreasing the level of US during treatment may modify the influence of the US on tumor growth. The results of our study suggest that this modified influence may be achieved by decreasing the intensity of the US or by switching to a pulsed US mode. When pulsed US is used, the intensity may be set higher than for continuous US, and treatment duration may be increased.

Despite this possible modified influence of lower levels of US, we continue to encourage caution in the use of US. We are convinced that US should not be used over an area where the presence of cancer has been positively established. Because there is a possibility that some patients may receive US prior to being diagnosed with cancer, using lower-level US or pulsed US and limiting the number of treatments should be considered. This approach may reduce the effects of US on unidentified cancerous tissue. The effectiveness of treatment with lower-level and pulsed US is being studied by numerous investigators. Clinicians should follow the literature as well as maintain careful records of their own results and use the lowest US levels possible to achieve satisfactory outcomes.

Conclusion

Ultrasound can stimulate tumor growth in mice. Reducing the amount of infused US energy by reducing intensity or by using pulsed US will result in a decrease in the amount of tumor growth as compared with higher-intensity continuous US. The amount of tumor growth may be related to the amount of infused US energy. Tumor metastases do not increase in vivo when using pulsed US on mice.

(*) Fort Dodge Laboratories Inc, 800 5th St NW, Fort Dodge, IA 50501.

([dagger]) Mobay Corp. Animal Health Div. 12707 W 63rd St, Shawnee, KS 66216.

([double dagger]) Delft Instruments Physical Medicine BV, PO Box 810, 2600 AV Delft, the Netherlands. Distributed in the United States by Henley International Inc, 10518 Kinghurst, Houston, TX 77099.

([sections]) Sylvac Fowler, 66 Rowe St, Newton, MA 02166.

([parallel]) Ohaus Corp, 29 Hanover Rd, Florham Park, NJ 07932.

References

[1] Lehmann JF, De Lateur BJ. Diathermy and superficial heat, laser, and cold therapy. In: Kottke FJ, Lehmann JF, eds. Krusen's Handbook of Physical Medicine and Rehabilitation. 4th ed. Philadelphia, Pa: WB Saunders Co; 1990:313-331.

[2] Ziskin MC, McDiarmid T, Michlovitz SL. Therapeutic ultrasound. In: Michlovitz SL. Thermal Agents in Rehabilitation. 2nd ed. Philadelphia, Pa: FA Davis Co; 1990: chap 7.

[3] Basford JR. Physical agents and biofeedback. In: DeLisa JA. Rehabilitation Medicine: Principles and Practice. Philadelphia, Pa: JB Lippincott Co; 1988:261-263.

[4] Sicard-Rosenbaum L, Lord D, Danoff, JV, et al. Effects of continuous therapeutic ultrasound on growth and metastasis of subcutaneous murine tumors. Phys Ther. 1995;75:3-13.

[5] Lejbkowicz F, Zwiran M, Salzberg S. The response of normal and malignant cells to ultrasound in vitro. Ultrasound Med Biol. 1993;19: 75-82.

[6] Patrick MR. Application of therapeutic pulsed ultrasound. Physiotherapy. 1978;64:103-104.

[7] Kramer JF. Ultrasound: evaluation of its mechanical and thermal effects, Arch Phys Med. 1955;65:223-227.

[8] Wong RA, Alexander RB, Puri RK. Rosenberg SA. In vivo proliferation of adoptively transferred tumor-infiltrating lymphocytes in mice. J Immunother. 1991:.10:120-130.

[9] Shrout FE, Fleiss J. Intraclass correlations: uses in assessing rater reliability. Psychol Bull 1979;86:420-428.

[10] Bartko JJ. On the methods and theory of reliability. J Nerv Ment Dis. 1976;163:307-315.

[11] Encyclopedia Americana. New York, NY: Americana Corp. 1977;10: 253-254.

[12] Andrews EJ, Bennett BT, Clark JD, et al, Report of the American Veterinary Medical Association Panel on Euthanasia. J Am Vet Med Assoc. 1993;202:229-249.

[13] Janik P, Briand P. Hartmann N. Effects of estrone-progesterone treatment on cell proliferation kinetics of hormone-dependent GR mouse mammary tumors. Cancer Res. 1975;35:3698-3704.

[14] Pizzo PA, Horowitz ME, Poplack DG, et al. Solid tumors of childhood In: Devita DV, Hellamn S, Rosenberg SA, eds. Cancer Principles and Practice of Oncology. 4th ed. Philadelphia, Pa: JB Lippincott Co. 1993:1738-1791.

[15] Siegel E, Goddard J, James E, et al. Cellular attachment as a sensitive indicator of the effects of diagnostic ultrasound exposure on cultured human cells. Radiology. 1979;133:175-179.

[16] Conger A, Siskin M. Wittels H. Ultrasonic effects on mammalian multicellular tumor spheroids. J Clin Ultrasound. 1981;9:167-174.

[17] Forytkova L, Hrazdira I, Mornstein V. Effect of ultrasound on DNA synthesis in tumor cells. Ultrasound Med Biol. 1995;21:585-592.