Background

High intensity ultrasound (HIUS) has been increasingly investigated as a possible tool in the treatment of many different tumor entities, e.g. cancers of prostate, kidney, liver, pancreaticobiliary and other intrabdominal malignancies [1,2,3]. While HIUS is still under clinical evaluation, previous studies indicate its potential to improve overall antitumoral activity regardless of chemotherapeutic applications [4]. Despite encouraging first clinical results [5], no studies have been conducted yet to assess HIUS’ possible application in the treatment of peritoneal metastases (PM). In all previous clinical applications, the HIUS beam was “focused” on a single spot in the body (High intensity focused ultrasound, HIFU). However, while PM usually covers a large surface, its depth is only minimal. This might be one reason as to why HIFU has not been considered for peritoneal applications. This study aims to modify the conventional HIFU to a “non-focused” HIUS approach and assess its potential in PM treatment. While HIUS is assumed to impact the peritoneum when used in the treatment of liver cancer, the validity of this assumption remains unclear [6]. Considering that any interaction with the peritoneum may be used in a therapeutic capacity, and inaccessibility remains one of the main difficulties in PM treatment, it seems astonishing that HIUS has never been investigated as a potential tool in PM treatment. PM is a common manifestation of advanced gastrointestinal and gynecological cancers, and affected patients usually have a very poor prognosis with median survival rates of only a few months [7]. Recent studies indicate that the combination of HIUS with intraperitoneal chemotherapy (IPC) could significantly increase drug penetration depths and therefore enhance the overall antitumoral effect, especially when applied with liposomal doxorubicin [8, 9]. While this effect has mostly been attributed to the rupture of liposomal doxorubicin [10], the results partially exceeded penetration levels observed in conventional chemotherapeutic solutions [8]. At the same time, no structural damage to the peritoneum was detected. Still, some authors have suggested that HIUS might affect the peritoneal surface when accidently applied during hepatocellular carcinoma treatment (HCC) [6]. To our knowledge, neither the application of HIUS and its potential, nor its possible side effects on the peritoneum have ever been systematically studied in the context of PM. In general surgery, HIUS is an established procedure predominantly used in the treatment of HCC [11, 12]. In a previous study, HIUS was assumed to cause local heat on the peritoneum, which could possibly induce peritoneal tissue destruction [6]. However, recent clinical evaluations indicate that HIUS might be safe for intraperitoneal use [13]. Knowing the antitumoral properties demonstrated by HIUS in HCC, it seems reasonable to assume similar effects in PM. Thus, with respect to its low invasiveness and absence of radiation, HIUS may potentially play an important role in future PM treatment. To evaluate the structural effects of HIUS on the peritoneum, we studied a well-established ex-vivo model in which we investigated peritoneal samples following HIUS application using light, fluorescence and electron microscopy.

Methods

No approval of the local board on animal welfare was required as the experiments were performed using commercially available tissue samples. A local animal supplier (Zerniki Wielkie, 55–020 Wroclaw, Poland) provided the fresh post-mortem swine peritoneum. This peritoneum was cut into proportional samples and placed into petri dishes. Then, the samples were covered with NaCl 0.9% until a layer of 6 mm covering the samples was attained (Fig. 1). HIUS (Sonopuls HD 2070, Bandelin, Berlin, Germany) at 70 W and 20 kHz was applied on the peritoneal tissue using a metal pen. The applied HIUS beam was not focused with high intensity, but rather spread from the tip of the metal pin to the periphery with continuously decreasing intensity. The tip of the pen was as close as 3 mm to the tissue. Each sample group included 3 peritoneal tissue samples and received either 30, 60, 120 or 300 s of HIUS treatment. The control group did not receive any HIUS exposure and was only placed in a petri dish for 300 s and covered by NaCl 0.9%. One sample of each group was subject to further analyses by means of light, fluorescence, or electron microscopy. Experiments were independently performed three times.

Fig. 1
figure 1

HIUS ex-vivo experiment on fresh swine peritoneum in 0.9% NaCl. The fresh full thick swine peritoneum sample is placed in a petri-dish filled with NaCl 0.9%. a Sonificator at 3 mm distance to the peritoneal sample. b Peritoneal sample. c NaCl 0.9%

Full size image

Light microscopy (LM) analysis

Peritoneal tissue was surgically removed and placed under a light microscope (Nikon Instruments Europe B.V. Amsterdam, Netherlands) to detect major structural changes. For samples that were later used in light microcopy, a temperature probe was placed at a 3 mm distance to the tip off the metal pen to measure a possible temperature increase.

Fluorescence microscopy (FM) analysis

The second group of samples was immediately frozen in liquid nitrogen to enable cryo sectioning (10 μm) of different areas of each specimen. To stain nuclei, sections were mounted with VectaShield containing 1.5 μg/ml 4′,6-diamidino-2-phenylindole (DAPI). Probes were analyzed using Nikon Eclipse 80i fluorescence microscope (Nikon Instruments Europe B.V. Amsterdam, Netherlands) and subperitoneal structural tissue damage was measured.

Electron microscopy (EM) analysis

A representative amount of the tissue sample was visualized using cryogenic scanning electron microscopy (cryo-SEM).

For this purpose, tissue samples were fixed in 2.5% glutaraldehyde solution in phosphate buffer (pH = 7.2) overnight. Following fixation, samples were cleaned in phosphate buffer, rinsed in ultrapure (filtered through 0.1um syringe filter) deionized water, mounted on cryo shuttle using OCT/colloidgraphite mixture and plunged in liquid nitrogen. Then, frozen samples were quickly transferred to the cryo-preparation chamber (Cryo Quorum PP3010T) and sputtered with a conductive platinum layer at -140C. In the next step, samples were transferred to the microscope chamber maintaining the same temperature of -140C (Auriga60, Zeiss) and observed at 2 kV of acceleration voltage using In Lens and SE2 secondary electron detectors.

Statistical analyses

Experiments were independently performed three times. Sigma Plot 12 (Systat Software Inc., California, USA) was used to perform statistical analysis. For analyses of independent groups, the Kruskal-Wallis One Way Analysis of Variance on Ranks was utilized. A p-value of < 0.05 was considered significant.

Results

Light microscopy (LM)

When probes were removed for further analysis, macroscopic changes on the peritoneal surface became detectable. Macroscopically, the peritoneum had become more whitish and presumably thicker. No visual signs of tissue tearing were detectable. There was no perforation in the peritoneal layer. The peritoneal surface became jelly-like after medium was removed for further preparation. Using a temperature probe, no temperature increase was detectable in the medium neither during the experiment and nor immediately after. After removal of the peritoneum for further light microscopy, the clear structural texture of the tissue was more whitish when compared to untreated samples. No clear signs of larger tissue disruptions were visible.

Fluorescence microscopy (FM)

Microscopic analysis of tissue samples showed a substantial structural difference compared to the control group. The superficial peritoneal layer of the samples showed signs of structural mechanical disintegration with ongoing HIUS duration (Figs. 2 and 3). The superficial peritoneal layer seemed to be disrupted into horizontal fibers. This effect seemed to increase with continuous treatment (Figs. 2 and 3). While in tissue samples with short HIUS exposure time, this structural disintegration was limited to some areas of the peritoneum, in probes treated for 120 s and longer these disintegrated areas fused and created several parallel lines of peritoneal filaments. Additionally, the subperitoneal muscle tissue was disrupted. However, disruptions were rather vertical than horizontal. Also, vertical disruption was observed to increase with longer exposure time to HIUS (Figs. 2 and 3). This increase in disruption size was significant from 48 +/− 18,5 μm to 153 +/− 34,5 μm (p < 0.01) (Fig. 4). Disruption depth into the subperitoneal tissue was measured (Fig. 4) and increased significantly from 494 +/− 54,1 μm to 765 +/− 96,7 μm (p < 0.01).

Fig. 2
figure 2

Tissue disruption following HIUS at different durations. Left side: changes on the peritoneum. Right side: changes of the subperitoneal tissue

Full size image
Fig. 3
figure 3

Microscopic model of HIUS’ effects on the peritoneal tissue. Left side: untreated tissue. Right side: HIUS treated peritoneum. a horizontal disruptions. b total disruption depth into subperitoneal tissue

Full size image
Fig. 4
figure 4

Left side: Size of horizontal disruptions following HIUS. Right side: Disruption depth into the subperitoneal tissue following HIUS. # = p > 0.05, * = p < 0.05, ** = p < 0.01

Full size image

Electron microscopy (EM)

The applied magnification was at a wide range between 500X and 5000X. Structural disintegration of the uppermost peritoneal layer was confirmed by EM in probes treated with (+) versus probes without (−) HIUS. The peritoneal surface was practically divided into bundles of fibers (Fig. 5). In contrast, untreated probes showed a compact and mostly smooth surface.

Fig. 5
figure 5

Light microscopy and Cryogenic scanning electron microscopy of peritoneal surface with (+) and without (−) HIUS, magnification level 1000X

Full size image

Discussion

While in the past few years, many improvements have been observed in chemotherapeutic regimens and new drug compositions, a significant amount of PM patients fail to respond to systemic and local treatments. This circumstance is mostly attributed to molecular mechanisms and limited drug distribution into the tumor [14]. Similar limitations have been observed in IPC during PM treatments [15, 16]. However, higher local drug disposition and increased tissue drug penetration is reported to enhance the anti-tumoral effect [17,18,19]. Attempts to improve tissue penetration rates by treating the peritoneal surface prior to chemotherapeutic application were mostly unsuccessful. For example, concepts like using an energy beam via radiation to prepare the peritoneal tissue for IPC have unfortunately not shown any improved penetration effects [20,21,22]. However, our data suggests that HIUS might be an easy, feasible additional feature in the treatment of PM. While our data is limited, and the study is preliminary in nature, our findings present the potential effects of HIUS on the peritoneum. In the future, these effects can be used in various applications. By creating very small tissue disruptions within the peritoneal surface, the transport of various particles through this main barrier is facilitated, resulting in significantly improved penetration rates of chemotherapeutic drugs. Some previous studies suggest these possible HIUS effects in combination with chemotherapy [8, 23]. However, these studies were primarily investigating drug tissue penetrations without emphasizing structural tissue changes, thus giving little explanation for this effect. A very recent study has, for the first time, analyzed drug penetration on the peritoneum following HIUS application, and the findings of this study indicate that penetration rates can be enhanced by more than threefold depending on the duration of the HIUS beam [24]. Since the effect of HIUS seems to show limitations in depth, it could be used for PM treatment during cytoreductive surgery to possibly disrupt the vascular network of single nodules. This concept is quite interesting since tumor nodules in PM are assumed to have a reduced blood supply compared to regular peritoneal tissue [25]. Other HIUS aspects. e.g. its role in the enhanced apoptosis of cancer cells has been recently discovered and requires further analysis [26]. Thus, HIUS potential for PM must be further investigated and warrants more studies to thoroughly investigate its potential. However, this present study offers important first insight of potential HIUS application to treat PM.

Conclusions

Our data indicate that HIUS creates disruptions in the peritoneal surface and its underlying tissues. In the subperitoneal tissue, HIUS application results in microbubble formation. Beside its direct effects on the peritoneum, these structural surface changes might also result in increased drug permeability.

To adequately assess HIUS’ efficacy as well as its therapeutic possibilities on the peritoneum, further studies are required.