Micron 84 (2016) 17–22
Contents lists available at ScienceDirect
Micron journal homepage: www.elsevier.com/locate/micron
Sinusoidal obstruction syndrome (SOS): A light and electron microscopy study in human liver C.P.H. Vreuls a,b,∗ , A. Driessen c,d , S.W.M. Olde Damink e,f,g , G.H. Koek h,i , H. Duimel j , M.A.J. van den Broek e,f , C.H.C. Dejong e,f , F. Braet j , E. Wisse h,i,j,k a
Department of Pathology, Maastricht University Medical Centre, Maastricht, The Netherlands Department of Pathology, Amphia Hospital, Breda, The Netherlands c Department of Pathology, Antwerp University Hospital, Edegem, Belgium d University of Antwerp, Antwerp, Belgium e Department of Surgery, Maastricht University Medical Centre, Maastricht, The Netherlands f NUTRIM School for Nutrition and Translational Research in Metabolism, and GROW: School for Oncology and Developmental Biology, Maastricht University Medical Centre, Maastricht, The Netherlands g Department of HPB Surgery, Royal Free Hospital, London, United Kingdom h Department of Internal Medicine, Division of Gastroenterology and Hepatology, Maastricht University Medical Centre, Maastricht, The Netherlands i NUTRIM, School for Nutrition, Toxocology and Metabolism, Maastricht, The Netherlands j Maastricht Multimodal Molecular Imaging Institute, Division of Nanoscopy, University of Maastricht, The Netherlands k School of Medical Sciences (Discipline of Anatomy and Histology) — The Bosch Institute and Australian Centre for Microscopy & Microanalysis, The University of Sydney, Australia b
a r t i c l e
i n f o
Article history: Received 16 November 2015 Received in revised form 25 January 2016 Accepted 9 February 2016 Available online 15 February 2016 Keywords: Sinusoidal obstruction syndrome Electron microscopy Light microscopy Liver Human Oxaliplatin
a b s t r a c t
Aims: Oxaliplatin is an important chemotherapeutic agent, used in the treatment of hepatic colorectal metastases, and known to induce the sinusoidal obstruction syndrome (SOS). Pathophysiological knowledge concerning SOS is based on a rat model. Therefore, the aim was to perform a comprehensive study of the features of human SOS, using both light microscopy (LM) and electron microscopy (EM). Methods and results: Included were all patients of whom wedge liver biopsies were collected during a partial hepatectomy for colorectal liver metastases, in a 4-year period. The wedge biopsy were perfusion fixated and processed for LM and EM. The SOS lesions were selected by LM and details were studied using EM. Material was available of 30 patients, of whom 28 patients received neo-adjuvant oxaliplatin. Eighteen (64%) of the 28 patients showed SOS lesions, based on microscopy. The lesions consisted of sinusoidal endothelial cell detachment from the space of Disse on EM. In the enlarged space of Disse a variable amount of erythrocytes were located. Conclusion: Sinusoidal endothelial cell detachment was present in human SOS, accompanied by enlargement of the space of Disse and erythrocytes in this area. These findings, originally described in a rat model, were now for the first time confirmed in human livers under clinically relevant settings. © 2016 Published by Elsevier Ltd.
1. Introduction Colorectal liver metastases can be treated surgically with curative intent, and this leads to a 5-year overall survival of around 50% (Aloia et al., 2006; Jones et al., 1987). Both patients with initially resectable and irresectable liver metastases often receive chemotherapy prior to liver surgery, including oxaliplatin.
∗ Corresponding author at: Amphia Hospital, Department of Clinical Pathology, PO Box 90158, 4800RK Breda, The Netherlands. E-mail address: [email protected] (C.P.H. Vreuls). http://dx.doi.org/10.1016/j.micron.2016.02.006 0968-4328/© 2016 Published by Elsevier Ltd.
Oxaliplatin-based chemotherapy aims to make liver metastases operable and surgery potentially still curative. In addition it can be determined whether tumor response takes place, which may inform patients and doctors about prognosis. The latter can be reached in about 15% of the patients with a 5-year survival of 33% (Nakano et al., 2008). Neo-adjuvant chemotherapy, however, is associated with considerable side-effects. Due to the toxic effects of oxaliplatin, up to 74% of patients develop sinusoidal obstruction syndrome (SOS) in the nontumor-bearing liver (Rubbia-Brandt, 2010; van den Broek et al., 2009, 2013). The clinical symptoms of advanced stages of SOS are portal hypertension, hepatomegaly, splenomegaly, ascites and hyper-
18
C.P.H. Vreuls et al. / Micron 84 (2016) 17–22
bilirubinemia (Jones et al., 1987). SOS can lead to prolonged hospital stay and higher morbidity after liver surgery (Nakano et al., 2008; Rubbia-Brandt, 2010). A fatal outcome may be the result of esophageal variceal bleeding, which can occur in every stage of the disease (van den Broek et al., 2009). At the time of surgery, the liver has a typical ‘blue’ appearance. The presence of SOS can be suspected clinically, but the gold standard of diagnosing is histopathology. In addition, there are several non-invasive detection methods, including elevated serum aspartate aminotransferase (ASAT) to platelet ratio index (APRI-score), increased spleen size determined by computed tomography and increased indocyanine green retention rate or hyaluronic acid levels (van den Broek et al., 2013). However, to-date, all these tests have only modest sensitivity and specificity for the diagnosis of SOS (Nakano et al., 2008; Soubrane et al., 2010; Narita et al., 2012; Overman et al., 2010). Current knowledge about the pathophysiological mechanism of SOS is based on an animal model, namely the monocrotaline rat model, described by DeLeve et al. (1999). The initial toxic damage is reported to occur to the sinusoidal endothelial cell, characterized by a decreased number of fenestrae and the development of gaps and larger holes through the sinusoidal endothelial cell, visualized by electron microscopy. This results in discontinuities of the sinusoidal endothelial wall, eventually with a loss of an intact sinusoidal endothelial cell lining and with blood elements present in the space of Disse. Furthermore, the surrounding hepatocytes show signs of necrosis and there is an actual loss in the number of Kupffer cells and stellate cells (DeLeve et al., 1999). The monocrotaline rat model is accepted to represent a reliable model to study toxin-induced SOS. However, the clinical symptoms of both the animal model and the human disease are not always comparable. Also following oxaliplatin administration, no induction of SOS occurs in the rat model, probably because of differences in uptake (Robinson et al., 2013). It is thus unclear whether monocrotaline-induced SOS in rats resembles oxaliplatininduced SOS in humans. Up to date, no extensive research has been performed on the exact pathophysiologic mechanism of oxaliplatin-induced SOS in humans and its comparability to the mechanisms in the monocrotaline rat model. One study briefly reported some of the features, as described above, in humans, but this was not the main focus of the paper (Rubbia-Brandt et al., 2004). Therefore, the aim of the present investigation was to perform a detailed and comprehensive investigation of the ultrastructural features seen in SOS due to oxaliplatin use in humans, using both light microscopy (LM) and electron microscopy (EM).
2. Materials and methods 2.1. Patients Patients undergoing a first partial hepatectomy for colorectal liver metastases at Maastricht University Medical Centre between September 2005 and September 2009 were included. Liver wedge biopsies were obtained in these patients at the start of laparotomy, prior to mobilization of the liver or intestines. The basic neo-adjuvant treatment schedule included six cycles of oxaliplatin. The actual number of treatment cycles differed per patient, depending on the radiological response according to Response Criteria In Solid Tumors (RECIST) (Husband et al., 2004), in combination with the patient’s physical condition and the sustained side effects (e.g., neurotoxicity). The study was performed in accordance with the ethical standards of the Declaration of Helsinki, and written informed consent was obtained from each patient.
Fig. 1. Scheme to depict the selection procedure for electron microscopy. A semithin section for light microscopy is made and used for orientation. Based on this material the site of the lesions is selected (A). The surrounding tissue is trimmed away, leading to the area of interest for both semi-thin (B) and ultra-thin (C), respectively light and transmission electron microscopy.
2.2. Methods At the operating theater, the wedge biopsies were injection perfused with glutaraldehyde fixative (1.5% in cacodylate buffer 0.067 mol/L), as described in detail earlier (Wisse et al., 2010). After perfusion, the paler and hardened parts of the biopsy were cut into slices (0.1 × 0.1 × 0.5 cm) for LM flat embedding, blocks (1 × 1 × 1 mm) for transmission electron microscopy (TEM) and strips (1 × 1 × 10 mm) for scanning electron microscopy (SEM). After 20 min in glutaraldehyde the specimens were washed by a buffer (cacodylate buffer 0.2 mol/L) and subsequently kept in osmium fixative (1% osmium tetroxide in phosphate buffer) for approximately 60 min. This was followed by a second wash in buffer (phosphate buffer 0.2 mol/L) to remove the osmium. Finally, the specimens were dehydrated by an increasing percentage of ethanol and embedded in Epon for LM and TEM and freeze fractured in 100% ethanol for SEM, as previously outlined (Wisse et al., 2010). Semi-thin and ultra-thin sections were cut, respectively 1 m and 60 nm thick. Semi-thin sections were used for LM and stained with toluidine-blue. Ultra-thin sections were used for TEM investigation and were stained with uranyl acetate and lead citrate (Wisse et al., 2010). For LM, semi-thin sections were used to view the presence and extent of the sinusoidal endothelial wall lesions and to select the area of interest for subsequent EM assessment. Briefly, when sinusoidal endothelial wall lesions were present in a certain region, this area was relocated in an ultramicrotome for making ultrathin sections for EM, enabling the collection of full comparative structural data. In the absence of sinusoidal endothelial wall lesions, the centrilobular zone was sampled extensively. To confirm that the correct region was sampled for EM, the ultra-thin section was preceded by an additional parallel semi-thin section (Fig. 1). All semi-thin sections were examined by two experienced pathologists, blinded from the clinical data.
C.P.H. Vreuls et al. / Micron 84 (2016) 17–22
19
Fig. 2. Light micrograph showing detachment of the sinusoidal endothelial cell from the space of Disse (grade 2). Next to the detachment of sinusoidal endothelial cells (SEC) there was a widening of the space of Disse. To a varying degree, the space of Disse contains erythrocytes (>). Magnification 40×.
For TEM, ultra-thin sections were used to study the site of the lesions in detail. Firstly, the architectural relation between sinusoidal endothelial cells, hepatocytes and the space of Disse was examined. If there was an enlarged space of Disse, the content was described and stated as no content, erythrocytes and/or blebs. Secondly, several features with respect to the sinusoidal endothelial cell were assessed, such as fenestrae, the presence of all organelles and an intact cytoplasm, stated as absent or present. Finally, SEM sections were used to study cellular topology, in particular the sinusoidal endothelial cell and the presence of fenestrae, and in addition, the content of the space of Disse. 2.3. Statistics For continuous data, statistical analyses were performed using a student T-test. Differences between dichotomous variables were calculated with a Fisher’s exact test. Analyses were performed using SPSS 15.0 for Windows (SPSS Inc., Chicago, Illinois, USA). A 64 days < 0.050 was considered significant. 3. Results 3.1. Patients LM, TEM and SEM material was available from 30 patients. Twenty-eight patients had received oxaliplatin treatment with a median number of 4.5 cycles (range 2–17). The median interval between the last administration of oxaliplatin and surgery was 62.0 days (range 26–1143 days). Two patients had not received chemotherapy, but the other pre-operative treatments were similar and therefore these patients served as adequate control. 3.2. SOS 3.2.1. Light microscopy Eighteen (64%) of the 28 patients showed sinusoidal lesions (Table 1). The lesions consisted of detachment of the sinusoidal endothelial cell from the space of Disse (Fig. 2). The sinusoidal endothelial cell detachment ranged from focally to pan-lobular involvement. Most commonly the centrilobular and mid-lobular area were affected. The time interval between the last administration of oxaliplatin and surgery ranged from 26 up to 1143 days and the sinusoidal endothelial cell detachment was observed in all patient samples for the given times.
Fig. 3. Transmission electron microscopy (TEM) image of the sinusoidal endothelial cell detachment from the space of Disse, in 2 different patients. The detached sinusoidal endothelial cells are still mutually connected and in the enlarged space of Disse multiple erythrocytes are located (* ). Magnification 2600× and 3400×.
Next to the detachment of sinusoidal endothelial cells there was a widening of the space of Disse, which in a varying degree contained packed erythrocytes (Table 1). The hepatocytes had a variation in diameter, containing al the organelles with presence of the bile canaliculi. The luminal side showed microvilli (Fig. 3). The median cumulative amount of oxaliplatin differed between both groups, sinusoidal endothelial cell detachment and absence of sinusoidal endothelial cell detachment, respectively 752.5 and 495 mg/m2 . Also the interval between the last administration of oxaliplatin and surgery differed in both groups, respectively 60.5 and 64 days (Table 1). Sinusoidal lesions were absent in the control group, namely the patients without oxaliplatin treatment.
3.3. Transmission and scanning electron microscopy At the site of the sinusoidal lesions, a detachment of the sinusoidal endothelial cell from the space of Disse (Figs. 3 and 4) was observed, despite the presence of collagen bundles in this space. The detached sinusoidal endothelial cells however were still mutually connected. In the enlarged space of Disse a varying amount of erythrocytes were located. The sinusoidal endothelial cells at the site of the sinusoidal lesions were further studied. The sinusoidal endothelial cell showed fenestrae, which were evenly distributed. Two patients showed hyperfenestration, i.e., the fenestrae were increased in quantity, but not in shape and size (Table 1). There was no rounding up of the sinusoidal endothelial cell. The sinusoidal endothelial cell contained all organelles, e.g., mitochondria and endoplasmic reticulum. At the site of the dam-
20
C.P.H. Vreuls et al. / Micron 84 (2016) 17–22
Table 1 Patient characteristics and morphologic features of all groups sinusoidal endothelial cell detachment, absence of sinusoidal endothelial cell detachment and control group.
Patient characteristics Sex, male, n Age, median ± SD Cumulative amount of OX, median ± SD, mg/m2 Interval between the last administration of OX and surgery, median ± SD, days Morphologic features Gap formation in sinusoidal endothelial wall, n Fenestrae, n Hyperfenestration, n Erythrocytes in space of Disse, n Giant mitochondria, n
Sinusoidal endothelial cell detachment (n = 18)
Absence of sinusoidal endothelial cell detachment (n = 10)
P
Control group (n = 2)
9 61.5 ± 13.2 752.5 ± 336.0 60.5 ± 299.1
7 66.0 ± 4.9 495 ± 218.7 64.0 ± 142.0
0.44 0.05 0.04a 0.49
2 70.0 ± 0 – –
0 18 (100%) 1 (6%) 15 (83%) 8 (44%)
0 10 (100%) 0 0 5 (50%)
– – – – 1.00
0 2 (100%) 1 (50%) 0 2 (100%)
Abbreviations: OX: oxaliplatin; SD: standard deviation. a Significant.
were also present in the space of Disse. Thirteen of the twenty eight patients showed a varying number of giant mitochondria (Table 1). The control group revealed a space of Disse with a normal width and no detachment of sinusoidal endothelial cells. The sinusoidal endothelial cells were well structured with presence of normal fenestrae between these cells. No collection of erythrocytes in the space of Disse was observed. Both patients showed giant mitochondria (Table 1). Giant mitochondria are larger in size than normal mitochondria and contain typical crystals and a highly rhythmical pattern of the cristae (Wisse et al., 2010). 4. Discussion
Fig. 4. Scanning electron microscopy (SEM) image of the sinusoidal endothelial cells detachment from the space of Disse, in 2 different patients. It is possible to view inside the sinusoid and focally the sinusoidal endothelial cells are detached and multiple erythrocytes are located in the space of Disse (* ). Magnification 2500× and 3400×.
aged sinusoids the extracellular matrix was present in the space of Disse, consisting of collagen bundles. The microvilli of the hepatocytes were less well-structured (i.e., irregular in shape) and/or lost. At the surface of the hepatocytes blebs were extending into the space of Disse. These blebs, which are an irregular bulge of the plasma membrane or swollen microvilli,
Light and electron microscopic examination of liver wedge biopsies of oxaliplatin-treated patients subjected to surgery for colorectal liver metastases, showed specific lesions in case of SOS. In those patients, the sinusoidal endothelial cell lining was detached and an enlarged space of Disse was observed that contained variable amounts of erythrocytes. These findings, originally described in the monocrotaline rat model (DeLeve et al., 1999), were now for the first time confirmed in human livers under clinically relevant settings. In the monocrotaline rat model, the sequelae of toxin-induced sinusoidal endothelial cell damage have been studied meticulously. The toxic agent (monocrotaline pyrrole) binds to actin, leading to depolymerisation of filamentous actin (F-actin) (Deleve et al., 2003). This depolymerisation is considered to be essential in the pathogenesis of SOS. The actin cytoskeleton is responsible for the cell shape, where F-actin depolymerisation leads to rounding up of the cells. The number of sinusoidal endothelial cell fenestrae decreases and gaps or larger wholes develop through the cells (DeLeve et al., 1999). In addition, F-actin depolymerisation leads to an increased expression of matrix metalloproteinase (MMP) 9, and to a lesser degree of MMP-2 (Deleve et al., 2003). The MMP9 (gelatinase B) is able to digest a number of extracellular matrix molecules, including collagen type IV (Nagase et al., 2006). In the monocrotaline rat model this is visible by rupture of the sinusoidal wall. Due to the gaps and larger holes in the SEC wall blood elements may invade the space of Disse. This damage progresses with loss of SEC and a diminished number of Kupffer cells and stellate cells. Erythrocytes accumulate in spaces formerly occupied by hepatic sinusoids. The surrounding hepatic parenchymal cells show necrosis. In the present study, the detachment of sinusoidal endothelial cells from the space of Disse was a prominent feature. Patients with sinusoidal endothelial cell detachment had received a larger median cumulative amount of oxaliplatin and had a tendency to a shorter interval between the last administration of oxali-
C.P.H. Vreuls et al. / Micron 84 (2016) 17–22
platin surgery, respectively 752.5 versus 495 mg/m2 and 60.5 and 64 days (Table 1), which is consistent with previous literature (Rubbia-Brandt et al., 2010). The SEC detachment varied from focally to pan-lobular involvement. If the detachment was only focally present, it was predominantly situated in the centrilobular area. This is in accordance with the known distribution of dilatation described in SOS, also starting from the centrilobular area (RubbiaBrandt et al., 2010). Independent of the time interval between the last administration of oxaliplatin and surgery, which ranged from 26 up to 1143 days, sinusoidal endothelial cell detachment was always present throughout the entire period. The presence of sinusoidal endothelial cell detachment even at 1143 days was an interesting finding. This finding assumes that restoration of the degraded extra cellular matrix is difficult. In this respect it would be interesting to investigate liver material, obtained at several timepoints during and after treatment with oxaliplatin. The SEC appeared to maintain their shape, as no rounding up was observed. The sinusoidal endothelial cells were mutually connected through intercellular junctions. Extensive gap formation in the sinusoidal endothelial cell or discontinuities in the sinusoidal endothelial lining were not observed. Focally there was hyperfenestration, but the fact that it was present in the study-group as well as the control group indicates the relative importance of this observation. In contrast to the monocrotaline rat model, no gaps were found in the sinusoidal endothelial cell or in the sinusoidal endothelial lining. A possible explanation for this was the long interval between the exposure to the toxic drug, oxaliplatin, and surgery, with a median time of 60.5 days (Table 1). If any gaps were present these could have been missed due to this delay. Nevertheless, there must have been some previous damage to the sinusoidal endothelial cell or sinusoidal endothelial lining since erythrocytes were found in the space of Disse. The presence of an intact sinusoidal endothelial lining, covering the enlarged space of Disse, suggested that restoration of the sinusoidal endothelial cell damage may occur very soon after exposure to the toxic product, oxaliplatin. In the animal model a period of deficiency of sinusoidal endothelial cells, which was followed with a recovery of these cells, has been described by DeLeve et al. (1999). The results of the present study suggest that recovery of the damage of the sinusoidal endothelial cell occurred shortly, at least within 26 days, after exposure to oxaliplatin, as the nature of the electron-microscopic abnormalities did not differ in function of the time interval between chemotherapy and surgery. Giant mitochondria, a typical feature of human liver parenchymal cells, were observed in patients with and without sinusoidal endothelial cell detachment (Table 1). According to literature, giant mitochondria are a diagnostic feature of recent and heavy alcoholism (Bruguera et al., 1977), but since we were not aware of the patients alcohol consumption this could not be correlated. The injection of the fixative through the sinusoids may focally lead to a high pressure in the sinusoids. This can cause artificial damage where the endothelial lining together with collagen fibers and bundles, processes of stellate cells and disrupted microvilli of parenchymal cells detached as a coherent complex from the surface of the parenchymal cells. The mechanical coherence of the endothelial lining with the underlying cellular structures and the extra-cellular matrix was apparently stronger than the inherent strength of the microvilli at the surface of the parenchymal cells. It is important to point out that there are two main differences compared with the described SOS lesions. Firstly, the SOS lesion showed detachment at the level of the space of Disse and the artificial damage showed disruption in the parenchymal cell itself. Secondly, the SOS lesions mainly contained erythrocytes in the subendothelial
21
Fig. 5. Transmission electron microscopy (TEM) image of a sinusoid (S) with artificial damage due to high pressure. The endothelial lining together with collagen fibers and bundles, processes of stellate cells and disrupted microvilli of parenchymal cells detach as a coherent complex from the surface of the parenchymal cells and gap formation occurs (A).
space, where the artificial damage contains only cell residues or the space is empty (Fig. 5). Recently, Robinson et al. (2013) have proposed a new animal model for SOS, namely a murine chronic model of oxaliplatininduced SOS. The mice were given intra-peritoneal chemotherapy treatment, including oxaliplatin, on a weekly basis for 5 weeks. Sinusoidal dilatation was studied by LM. The authors stated that their model is better comparable with the human situation since they use the same toxic agent and the drug exposure is chronic. It would be of great value obtaining EM material from this new animal model and there with enabling comparison with the features described in this present human study and the monocrotaline rat model. To expand our insight in the pathophysiology of SOS in humans, it would be interesting to study several moments in one and the same human liver after administration of oxaliplatin, at the LM and EM level. The most obvious approach would be through obtaining several serial liver needle biopsies. Especially, since an easily applicable method for EM fixation of liver needle biopsies has recently been described (Vreuls et al., 2013) this seems a realistic perspective. This technique will allow us to study the sinusoids and sinusoidal endothelial cell in detail. Unfortunately, it is until now ethically not justified to obtain a liver needle biopsy in oxaliplatintreated patients shortly after treatment. We are planning to study the effect of oxaliplatin-based chemotherapy in biopsies obtained in patients who had undergone repeated liver surgery for recurrent colorectal liver metastases. This project might give us more information about the pathophysiology and reversibility of the lesions. References Aloia, T., Sebagh, M., Plasse, M., Karam, V., Levi, F., Giacchetti, S., Azoulay, D., et al., 2006. Liver histology and surgical outcomes after preoperative chemotherapy with fluorouracil plus oxaliplatin in colorectal cancer liver metastases. J. Clin. Oncol. 24, 4983–4990. Bruguera, M., Bertran, A., Bombi, J.A., Rodes, J., 1977. Giant mitochondria in hepatocytes: a diagnostic hint for alcoholic liver disease. Gastroenterology 73, 1383–1387. DeLeve, L.D., McCuskey, R.S., Wang, X., Hu, L., McCuskey, M.K., Epstein, R.B., Kanel, G.C., 1999. Characterization of a reproducible rat model of hepatic veno-occlusive disease. Hepatology 29, 1779–1791. Deleve, L.D., Wang, X., Tsai, J., Kanel, G., Strasberg, S., Tokes, Z.A., 2003. Sinusoidal obstruction syndrome (veno-occlusive disease) in the rat is prevented by matrix metalloproteinase inhibition. Gastroenterology 125, 882–890. Husband, J.E., Schwartz, L.H., Spencer, J., Ollivier, L., King, D.M., Johnson, R., Reznek, R., 2004. Evaluation of the response to treatment of solid tumors — a consensus statement of the International Cancer Imaging Society. Br. J. Cancer 90, 2256–2260.
22
C.P.H. Vreuls et al. / Micron 84 (2016) 17–22
Jones, R.J., Lee, K.S., Beschorner, W.E., Vogel, V.G., Grochow, L.B., Braine, H.G., Vogelsang, G.B., et al., 1987. Venoocclusive disease of the liver following bone marrow transplantation. Transplantation 44, 778–783. Nagase, H., Visse, R., Murphy, G., 2006. Structure and function of matrix metalloproteinases and TIMPs. Cardiovasc. Res. 69, 562–573. Nakano, H., Oussoultzoglou, E., Rosso, E., Casnedi, S., Chenard-Neu, M.P., Dufour, P., Bachellier, P., et al., 2008. Sinusoidal injury increases morbidity after major hepatectomy in patients with colorectal liver metastases receiving preoperative chemotherapy. Ann. Surg. 247, 118–124. Narita, M., Oussoultzoglou, E., Fuchshuber, P., Pessaux, P., Chenard, M.P., Rosso, E., Nobili, C., et al., 2012. What is a safe future liver remnant size in patients undergoing major hepatectomy for colorectal liver metastases and treated by intensive preoperative chemotherapy? Ann. Surg. Oncol. (e-pub ahead of print). Overman, M.J., Maru, D.M., Charnsangavej, C., Loyer, E.M., Wang, H., Pathak, P., Eng, C., et al., 2010. Oxaliplatin-mediated increase in spleen size as a biomarker for the development of hepatic sinusoidal injury. J. Clin. Oncol. 28, 2549–2555. Robinson, S.M., Mann, J., Vasilaki, A., Mathers, J., Burt, A.D., Oakley, F., White, S.A., et al., 2013. Pathogenesis of folfox induced sinusoidal obstruction syndrome in a murine chemotherapy model. J. Hepatol. Rubbia-Brandt, L., 2010. Sinusoidal obstruction syndrome. Clin. Liver Dis. 14, 651–668. Rubbia-Brandt, L., Audard, V., Sartoretti, P., Roth, A.D., Brezault, C., Le Charpentier, M., Dousset, B., et al., 2004. Severe hepatic sinusoidal obstruction associated with oxaliplatin-based chemotherapy in patients with metastatic colorectal cancer. Ann. Oncol. 15, 460–466.
Rubbia-Brandt, L., Lauwers, G.Y., Wang, H., Majno, P.E., Tanabe, K., Zhu, A.X., Brezault, C., et al., 2010. Sinusoidal obstruction syndrome and nodular regenerative hyperplasia are frequent oxaliplatin-associated liver lesions and partially prevented by bevacizumab in patients with hepatic colorectal metastasis. Histopathology 56, 430–439. Soubrane, O., Brouquet, A., Zalinski, S., Terris, B., Brezault, C., Mallet, V., Goldwasser, F., et al., 2010. Predicting high grade lesions of sinusoidal obstruction syndrome related to oxaliplatin-based chemotherapy for colorectal liver metastases: correlation with post-hepatectomy outcome. Ann. Surg. 251, 454–460. van den Broek, M.A., Olde Damink, S.W., Driessen, A., Dejong, C.H., Bemelmans, M.H., 2009. Nodular regenerative hyperplasia secondary to neoadjuvant chemotherapy for colorectal liver metastases. Case Report Med. 2009, 457975. van den Broek, M.A., Vreuls, C.P., Winstanley, A., Jansen, R.L., van Bijnen, A.A., Dello, S.A., Bemelmans, M.H., et al., 2013. Hyaluronic acid as a marker of hepatic sinusoidal obstruction syndrome secondary to oxaliplatin-based chemotherapy in patients with colorectal liver metastases. Ann. Surg. Oncol. 20, 1462–1469. Vreuls, C., Wisse, E., Duimel, H., Stevens, K., Verheyen, F., Braet, F., Driessen, A., et al., 2013. Jet-fixation: a novel method to improve microscopy of human liver needle biopsies. Hepatology. Wisse, E., Braet, F., Duimel, H., Vreuls, C., Koek, G., Olde Damink, S.W., van den Broek, M.A., et al., 2010. Fixation methods for electron microscopy of human and other liver. World J. Gastroenterol. 16, 2851–2866.
Contents lists available at ScienceDirect
Micron journal homepage: www.elsevier.com/locate/micron
Sinusoidal obstruction syndrome (SOS): A light and electron microscopy study in human liver C.P.H. Vreuls a,b,∗ , A. Driessen c,d , S.W.M. Olde Damink e,f,g , G.H. Koek h,i , H. Duimel j , M.A.J. van den Broek e,f , C.H.C. Dejong e,f , F. Braet j , E. Wisse h,i,j,k a
Department of Pathology, Maastricht University Medical Centre, Maastricht, The Netherlands Department of Pathology, Amphia Hospital, Breda, The Netherlands c Department of Pathology, Antwerp University Hospital, Edegem, Belgium d University of Antwerp, Antwerp, Belgium e Department of Surgery, Maastricht University Medical Centre, Maastricht, The Netherlands f NUTRIM School for Nutrition and Translational Research in Metabolism, and GROW: School for Oncology and Developmental Biology, Maastricht University Medical Centre, Maastricht, The Netherlands g Department of HPB Surgery, Royal Free Hospital, London, United Kingdom h Department of Internal Medicine, Division of Gastroenterology and Hepatology, Maastricht University Medical Centre, Maastricht, The Netherlands i NUTRIM, School for Nutrition, Toxocology and Metabolism, Maastricht, The Netherlands j Maastricht Multimodal Molecular Imaging Institute, Division of Nanoscopy, University of Maastricht, The Netherlands k School of Medical Sciences (Discipline of Anatomy and Histology) — The Bosch Institute and Australian Centre for Microscopy & Microanalysis, The University of Sydney, Australia b
a r t i c l e
i n f o
Article history: Received 16 November 2015 Received in revised form 25 January 2016 Accepted 9 February 2016 Available online 15 February 2016 Keywords: Sinusoidal obstruction syndrome Electron microscopy Light microscopy Liver Human Oxaliplatin
a b s t r a c t
Aims: Oxaliplatin is an important chemotherapeutic agent, used in the treatment of hepatic colorectal metastases, and known to induce the sinusoidal obstruction syndrome (SOS). Pathophysiological knowledge concerning SOS is based on a rat model. Therefore, the aim was to perform a comprehensive study of the features of human SOS, using both light microscopy (LM) and electron microscopy (EM). Methods and results: Included were all patients of whom wedge liver biopsies were collected during a partial hepatectomy for colorectal liver metastases, in a 4-year period. The wedge biopsy were perfusion fixated and processed for LM and EM. The SOS lesions were selected by LM and details were studied using EM. Material was available of 30 patients, of whom 28 patients received neo-adjuvant oxaliplatin. Eighteen (64%) of the 28 patients showed SOS lesions, based on microscopy. The lesions consisted of sinusoidal endothelial cell detachment from the space of Disse on EM. In the enlarged space of Disse a variable amount of erythrocytes were located. Conclusion: Sinusoidal endothelial cell detachment was present in human SOS, accompanied by enlargement of the space of Disse and erythrocytes in this area. These findings, originally described in a rat model, were now for the first time confirmed in human livers under clinically relevant settings. © 2016 Published by Elsevier Ltd.
1. Introduction Colorectal liver metastases can be treated surgically with curative intent, and this leads to a 5-year overall survival of around 50% (Aloia et al., 2006; Jones et al., 1987). Both patients with initially resectable and irresectable liver metastases often receive chemotherapy prior to liver surgery, including oxaliplatin.
∗ Corresponding author at: Amphia Hospital, Department of Clinical Pathology, PO Box 90158, 4800RK Breda, The Netherlands. E-mail address: [email protected] (C.P.H. Vreuls). http://dx.doi.org/10.1016/j.micron.2016.02.006 0968-4328/© 2016 Published by Elsevier Ltd.
Oxaliplatin-based chemotherapy aims to make liver metastases operable and surgery potentially still curative. In addition it can be determined whether tumor response takes place, which may inform patients and doctors about prognosis. The latter can be reached in about 15% of the patients with a 5-year survival of 33% (Nakano et al., 2008). Neo-adjuvant chemotherapy, however, is associated with considerable side-effects. Due to the toxic effects of oxaliplatin, up to 74% of patients develop sinusoidal obstruction syndrome (SOS) in the nontumor-bearing liver (Rubbia-Brandt, 2010; van den Broek et al., 2009, 2013). The clinical symptoms of advanced stages of SOS are portal hypertension, hepatomegaly, splenomegaly, ascites and hyper-
18
C.P.H. Vreuls et al. / Micron 84 (2016) 17–22
bilirubinemia (Jones et al., 1987). SOS can lead to prolonged hospital stay and higher morbidity after liver surgery (Nakano et al., 2008; Rubbia-Brandt, 2010). A fatal outcome may be the result of esophageal variceal bleeding, which can occur in every stage of the disease (van den Broek et al., 2009). At the time of surgery, the liver has a typical ‘blue’ appearance. The presence of SOS can be suspected clinically, but the gold standard of diagnosing is histopathology. In addition, there are several non-invasive detection methods, including elevated serum aspartate aminotransferase (ASAT) to platelet ratio index (APRI-score), increased spleen size determined by computed tomography and increased indocyanine green retention rate or hyaluronic acid levels (van den Broek et al., 2013). However, to-date, all these tests have only modest sensitivity and specificity for the diagnosis of SOS (Nakano et al., 2008; Soubrane et al., 2010; Narita et al., 2012; Overman et al., 2010). Current knowledge about the pathophysiological mechanism of SOS is based on an animal model, namely the monocrotaline rat model, described by DeLeve et al. (1999). The initial toxic damage is reported to occur to the sinusoidal endothelial cell, characterized by a decreased number of fenestrae and the development of gaps and larger holes through the sinusoidal endothelial cell, visualized by electron microscopy. This results in discontinuities of the sinusoidal endothelial wall, eventually with a loss of an intact sinusoidal endothelial cell lining and with blood elements present in the space of Disse. Furthermore, the surrounding hepatocytes show signs of necrosis and there is an actual loss in the number of Kupffer cells and stellate cells (DeLeve et al., 1999). The monocrotaline rat model is accepted to represent a reliable model to study toxin-induced SOS. However, the clinical symptoms of both the animal model and the human disease are not always comparable. Also following oxaliplatin administration, no induction of SOS occurs in the rat model, probably because of differences in uptake (Robinson et al., 2013). It is thus unclear whether monocrotaline-induced SOS in rats resembles oxaliplatininduced SOS in humans. Up to date, no extensive research has been performed on the exact pathophysiologic mechanism of oxaliplatin-induced SOS in humans and its comparability to the mechanisms in the monocrotaline rat model. One study briefly reported some of the features, as described above, in humans, but this was not the main focus of the paper (Rubbia-Brandt et al., 2004). Therefore, the aim of the present investigation was to perform a detailed and comprehensive investigation of the ultrastructural features seen in SOS due to oxaliplatin use in humans, using both light microscopy (LM) and electron microscopy (EM).
2. Materials and methods 2.1. Patients Patients undergoing a first partial hepatectomy for colorectal liver metastases at Maastricht University Medical Centre between September 2005 and September 2009 were included. Liver wedge biopsies were obtained in these patients at the start of laparotomy, prior to mobilization of the liver or intestines. The basic neo-adjuvant treatment schedule included six cycles of oxaliplatin. The actual number of treatment cycles differed per patient, depending on the radiological response according to Response Criteria In Solid Tumors (RECIST) (Husband et al., 2004), in combination with the patient’s physical condition and the sustained side effects (e.g., neurotoxicity). The study was performed in accordance with the ethical standards of the Declaration of Helsinki, and written informed consent was obtained from each patient.
Fig. 1. Scheme to depict the selection procedure for electron microscopy. A semithin section for light microscopy is made and used for orientation. Based on this material the site of the lesions is selected (A). The surrounding tissue is trimmed away, leading to the area of interest for both semi-thin (B) and ultra-thin (C), respectively light and transmission electron microscopy.
2.2. Methods At the operating theater, the wedge biopsies were injection perfused with glutaraldehyde fixative (1.5% in cacodylate buffer 0.067 mol/L), as described in detail earlier (Wisse et al., 2010). After perfusion, the paler and hardened parts of the biopsy were cut into slices (0.1 × 0.1 × 0.5 cm) for LM flat embedding, blocks (1 × 1 × 1 mm) for transmission electron microscopy (TEM) and strips (1 × 1 × 10 mm) for scanning electron microscopy (SEM). After 20 min in glutaraldehyde the specimens were washed by a buffer (cacodylate buffer 0.2 mol/L) and subsequently kept in osmium fixative (1% osmium tetroxide in phosphate buffer) for approximately 60 min. This was followed by a second wash in buffer (phosphate buffer 0.2 mol/L) to remove the osmium. Finally, the specimens were dehydrated by an increasing percentage of ethanol and embedded in Epon for LM and TEM and freeze fractured in 100% ethanol for SEM, as previously outlined (Wisse et al., 2010). Semi-thin and ultra-thin sections were cut, respectively 1 m and 60 nm thick. Semi-thin sections were used for LM and stained with toluidine-blue. Ultra-thin sections were used for TEM investigation and were stained with uranyl acetate and lead citrate (Wisse et al., 2010). For LM, semi-thin sections were used to view the presence and extent of the sinusoidal endothelial wall lesions and to select the area of interest for subsequent EM assessment. Briefly, when sinusoidal endothelial wall lesions were present in a certain region, this area was relocated in an ultramicrotome for making ultrathin sections for EM, enabling the collection of full comparative structural data. In the absence of sinusoidal endothelial wall lesions, the centrilobular zone was sampled extensively. To confirm that the correct region was sampled for EM, the ultra-thin section was preceded by an additional parallel semi-thin section (Fig. 1). All semi-thin sections were examined by two experienced pathologists, blinded from the clinical data.
C.P.H. Vreuls et al. / Micron 84 (2016) 17–22
19
Fig. 2. Light micrograph showing detachment of the sinusoidal endothelial cell from the space of Disse (grade 2). Next to the detachment of sinusoidal endothelial cells (SEC) there was a widening of the space of Disse. To a varying degree, the space of Disse contains erythrocytes (>). Magnification 40×.
For TEM, ultra-thin sections were used to study the site of the lesions in detail. Firstly, the architectural relation between sinusoidal endothelial cells, hepatocytes and the space of Disse was examined. If there was an enlarged space of Disse, the content was described and stated as no content, erythrocytes and/or blebs. Secondly, several features with respect to the sinusoidal endothelial cell were assessed, such as fenestrae, the presence of all organelles and an intact cytoplasm, stated as absent or present. Finally, SEM sections were used to study cellular topology, in particular the sinusoidal endothelial cell and the presence of fenestrae, and in addition, the content of the space of Disse. 2.3. Statistics For continuous data, statistical analyses were performed using a student T-test. Differences between dichotomous variables were calculated with a Fisher’s exact test. Analyses were performed using SPSS 15.0 for Windows (SPSS Inc., Chicago, Illinois, USA). A 64 days < 0.050 was considered significant. 3. Results 3.1. Patients LM, TEM and SEM material was available from 30 patients. Twenty-eight patients had received oxaliplatin treatment with a median number of 4.5 cycles (range 2–17). The median interval between the last administration of oxaliplatin and surgery was 62.0 days (range 26–1143 days). Two patients had not received chemotherapy, but the other pre-operative treatments were similar and therefore these patients served as adequate control. 3.2. SOS 3.2.1. Light microscopy Eighteen (64%) of the 28 patients showed sinusoidal lesions (Table 1). The lesions consisted of detachment of the sinusoidal endothelial cell from the space of Disse (Fig. 2). The sinusoidal endothelial cell detachment ranged from focally to pan-lobular involvement. Most commonly the centrilobular and mid-lobular area were affected. The time interval between the last administration of oxaliplatin and surgery ranged from 26 up to 1143 days and the sinusoidal endothelial cell detachment was observed in all patient samples for the given times.
Fig. 3. Transmission electron microscopy (TEM) image of the sinusoidal endothelial cell detachment from the space of Disse, in 2 different patients. The detached sinusoidal endothelial cells are still mutually connected and in the enlarged space of Disse multiple erythrocytes are located (* ). Magnification 2600× and 3400×.
Next to the detachment of sinusoidal endothelial cells there was a widening of the space of Disse, which in a varying degree contained packed erythrocytes (Table 1). The hepatocytes had a variation in diameter, containing al the organelles with presence of the bile canaliculi. The luminal side showed microvilli (Fig. 3). The median cumulative amount of oxaliplatin differed between both groups, sinusoidal endothelial cell detachment and absence of sinusoidal endothelial cell detachment, respectively 752.5 and 495 mg/m2 . Also the interval between the last administration of oxaliplatin and surgery differed in both groups, respectively 60.5 and 64 days (Table 1). Sinusoidal lesions were absent in the control group, namely the patients without oxaliplatin treatment.
3.3. Transmission and scanning electron microscopy At the site of the sinusoidal lesions, a detachment of the sinusoidal endothelial cell from the space of Disse (Figs. 3 and 4) was observed, despite the presence of collagen bundles in this space. The detached sinusoidal endothelial cells however were still mutually connected. In the enlarged space of Disse a varying amount of erythrocytes were located. The sinusoidal endothelial cells at the site of the sinusoidal lesions were further studied. The sinusoidal endothelial cell showed fenestrae, which were evenly distributed. Two patients showed hyperfenestration, i.e., the fenestrae were increased in quantity, but not in shape and size (Table 1). There was no rounding up of the sinusoidal endothelial cell. The sinusoidal endothelial cell contained all organelles, e.g., mitochondria and endoplasmic reticulum. At the site of the dam-
20
C.P.H. Vreuls et al. / Micron 84 (2016) 17–22
Table 1 Patient characteristics and morphologic features of all groups sinusoidal endothelial cell detachment, absence of sinusoidal endothelial cell detachment and control group.
Patient characteristics Sex, male, n Age, median ± SD Cumulative amount of OX, median ± SD, mg/m2 Interval between the last administration of OX and surgery, median ± SD, days Morphologic features Gap formation in sinusoidal endothelial wall, n Fenestrae, n Hyperfenestration, n Erythrocytes in space of Disse, n Giant mitochondria, n
Sinusoidal endothelial cell detachment (n = 18)
Absence of sinusoidal endothelial cell detachment (n = 10)
P
Control group (n = 2)
9 61.5 ± 13.2 752.5 ± 336.0 60.5 ± 299.1
7 66.0 ± 4.9 495 ± 218.7 64.0 ± 142.0
0.44 0.05 0.04a 0.49
2 70.0 ± 0 – –
0 18 (100%) 1 (6%) 15 (83%) 8 (44%)
0 10 (100%) 0 0 5 (50%)
– – – – 1.00
0 2 (100%) 1 (50%) 0 2 (100%)
Abbreviations: OX: oxaliplatin; SD: standard deviation. a Significant.
were also present in the space of Disse. Thirteen of the twenty eight patients showed a varying number of giant mitochondria (Table 1). The control group revealed a space of Disse with a normal width and no detachment of sinusoidal endothelial cells. The sinusoidal endothelial cells were well structured with presence of normal fenestrae between these cells. No collection of erythrocytes in the space of Disse was observed. Both patients showed giant mitochondria (Table 1). Giant mitochondria are larger in size than normal mitochondria and contain typical crystals and a highly rhythmical pattern of the cristae (Wisse et al., 2010). 4. Discussion
Fig. 4. Scanning electron microscopy (SEM) image of the sinusoidal endothelial cells detachment from the space of Disse, in 2 different patients. It is possible to view inside the sinusoid and focally the sinusoidal endothelial cells are detached and multiple erythrocytes are located in the space of Disse (* ). Magnification 2500× and 3400×.
aged sinusoids the extracellular matrix was present in the space of Disse, consisting of collagen bundles. The microvilli of the hepatocytes were less well-structured (i.e., irregular in shape) and/or lost. At the surface of the hepatocytes blebs were extending into the space of Disse. These blebs, which are an irregular bulge of the plasma membrane or swollen microvilli,
Light and electron microscopic examination of liver wedge biopsies of oxaliplatin-treated patients subjected to surgery for colorectal liver metastases, showed specific lesions in case of SOS. In those patients, the sinusoidal endothelial cell lining was detached and an enlarged space of Disse was observed that contained variable amounts of erythrocytes. These findings, originally described in the monocrotaline rat model (DeLeve et al., 1999), were now for the first time confirmed in human livers under clinically relevant settings. In the monocrotaline rat model, the sequelae of toxin-induced sinusoidal endothelial cell damage have been studied meticulously. The toxic agent (monocrotaline pyrrole) binds to actin, leading to depolymerisation of filamentous actin (F-actin) (Deleve et al., 2003). This depolymerisation is considered to be essential in the pathogenesis of SOS. The actin cytoskeleton is responsible for the cell shape, where F-actin depolymerisation leads to rounding up of the cells. The number of sinusoidal endothelial cell fenestrae decreases and gaps or larger wholes develop through the cells (DeLeve et al., 1999). In addition, F-actin depolymerisation leads to an increased expression of matrix metalloproteinase (MMP) 9, and to a lesser degree of MMP-2 (Deleve et al., 2003). The MMP9 (gelatinase B) is able to digest a number of extracellular matrix molecules, including collagen type IV (Nagase et al., 2006). In the monocrotaline rat model this is visible by rupture of the sinusoidal wall. Due to the gaps and larger holes in the SEC wall blood elements may invade the space of Disse. This damage progresses with loss of SEC and a diminished number of Kupffer cells and stellate cells. Erythrocytes accumulate in spaces formerly occupied by hepatic sinusoids. The surrounding hepatic parenchymal cells show necrosis. In the present study, the detachment of sinusoidal endothelial cells from the space of Disse was a prominent feature. Patients with sinusoidal endothelial cell detachment had received a larger median cumulative amount of oxaliplatin and had a tendency to a shorter interval between the last administration of oxali-
C.P.H. Vreuls et al. / Micron 84 (2016) 17–22
platin surgery, respectively 752.5 versus 495 mg/m2 and 60.5 and 64 days (Table 1), which is consistent with previous literature (Rubbia-Brandt et al., 2010). The SEC detachment varied from focally to pan-lobular involvement. If the detachment was only focally present, it was predominantly situated in the centrilobular area. This is in accordance with the known distribution of dilatation described in SOS, also starting from the centrilobular area (RubbiaBrandt et al., 2010). Independent of the time interval between the last administration of oxaliplatin and surgery, which ranged from 26 up to 1143 days, sinusoidal endothelial cell detachment was always present throughout the entire period. The presence of sinusoidal endothelial cell detachment even at 1143 days was an interesting finding. This finding assumes that restoration of the degraded extra cellular matrix is difficult. In this respect it would be interesting to investigate liver material, obtained at several timepoints during and after treatment with oxaliplatin. The SEC appeared to maintain their shape, as no rounding up was observed. The sinusoidal endothelial cells were mutually connected through intercellular junctions. Extensive gap formation in the sinusoidal endothelial cell or discontinuities in the sinusoidal endothelial lining were not observed. Focally there was hyperfenestration, but the fact that it was present in the study-group as well as the control group indicates the relative importance of this observation. In contrast to the monocrotaline rat model, no gaps were found in the sinusoidal endothelial cell or in the sinusoidal endothelial lining. A possible explanation for this was the long interval between the exposure to the toxic drug, oxaliplatin, and surgery, with a median time of 60.5 days (Table 1). If any gaps were present these could have been missed due to this delay. Nevertheless, there must have been some previous damage to the sinusoidal endothelial cell or sinusoidal endothelial lining since erythrocytes were found in the space of Disse. The presence of an intact sinusoidal endothelial lining, covering the enlarged space of Disse, suggested that restoration of the sinusoidal endothelial cell damage may occur very soon after exposure to the toxic product, oxaliplatin. In the animal model a period of deficiency of sinusoidal endothelial cells, which was followed with a recovery of these cells, has been described by DeLeve et al. (1999). The results of the present study suggest that recovery of the damage of the sinusoidal endothelial cell occurred shortly, at least within 26 days, after exposure to oxaliplatin, as the nature of the electron-microscopic abnormalities did not differ in function of the time interval between chemotherapy and surgery. Giant mitochondria, a typical feature of human liver parenchymal cells, were observed in patients with and without sinusoidal endothelial cell detachment (Table 1). According to literature, giant mitochondria are a diagnostic feature of recent and heavy alcoholism (Bruguera et al., 1977), but since we were not aware of the patients alcohol consumption this could not be correlated. The injection of the fixative through the sinusoids may focally lead to a high pressure in the sinusoids. This can cause artificial damage where the endothelial lining together with collagen fibers and bundles, processes of stellate cells and disrupted microvilli of parenchymal cells detached as a coherent complex from the surface of the parenchymal cells. The mechanical coherence of the endothelial lining with the underlying cellular structures and the extra-cellular matrix was apparently stronger than the inherent strength of the microvilli at the surface of the parenchymal cells. It is important to point out that there are two main differences compared with the described SOS lesions. Firstly, the SOS lesion showed detachment at the level of the space of Disse and the artificial damage showed disruption in the parenchymal cell itself. Secondly, the SOS lesions mainly contained erythrocytes in the subendothelial
21
Fig. 5. Transmission electron microscopy (TEM) image of a sinusoid (S) with artificial damage due to high pressure. The endothelial lining together with collagen fibers and bundles, processes of stellate cells and disrupted microvilli of parenchymal cells detach as a coherent complex from the surface of the parenchymal cells and gap formation occurs (A).
space, where the artificial damage contains only cell residues or the space is empty (Fig. 5). Recently, Robinson et al. (2013) have proposed a new animal model for SOS, namely a murine chronic model of oxaliplatininduced SOS. The mice were given intra-peritoneal chemotherapy treatment, including oxaliplatin, on a weekly basis for 5 weeks. Sinusoidal dilatation was studied by LM. The authors stated that their model is better comparable with the human situation since they use the same toxic agent and the drug exposure is chronic. It would be of great value obtaining EM material from this new animal model and there with enabling comparison with the features described in this present human study and the monocrotaline rat model. To expand our insight in the pathophysiology of SOS in humans, it would be interesting to study several moments in one and the same human liver after administration of oxaliplatin, at the LM and EM level. The most obvious approach would be through obtaining several serial liver needle biopsies. Especially, since an easily applicable method for EM fixation of liver needle biopsies has recently been described (Vreuls et al., 2013) this seems a realistic perspective. This technique will allow us to study the sinusoids and sinusoidal endothelial cell in detail. Unfortunately, it is until now ethically not justified to obtain a liver needle biopsy in oxaliplatintreated patients shortly after treatment. We are planning to study the effect of oxaliplatin-based chemotherapy in biopsies obtained in patients who had undergone repeated liver surgery for recurrent colorectal liver metastases. This project might give us more information about the pathophysiology and reversibility of the lesions. References Aloia, T., Sebagh, M., Plasse, M., Karam, V., Levi, F., Giacchetti, S., Azoulay, D., et al., 2006. Liver histology and surgical outcomes after preoperative chemotherapy with fluorouracil plus oxaliplatin in colorectal cancer liver metastases. J. Clin. Oncol. 24, 4983–4990. Bruguera, M., Bertran, A., Bombi, J.A., Rodes, J., 1977. Giant mitochondria in hepatocytes: a diagnostic hint for alcoholic liver disease. Gastroenterology 73, 1383–1387. DeLeve, L.D., McCuskey, R.S., Wang, X., Hu, L., McCuskey, M.K., Epstein, R.B., Kanel, G.C., 1999. Characterization of a reproducible rat model of hepatic veno-occlusive disease. Hepatology 29, 1779–1791. Deleve, L.D., Wang, X., Tsai, J., Kanel, G., Strasberg, S., Tokes, Z.A., 2003. Sinusoidal obstruction syndrome (veno-occlusive disease) in the rat is prevented by matrix metalloproteinase inhibition. Gastroenterology 125, 882–890. Husband, J.E., Schwartz, L.H., Spencer, J., Ollivier, L., King, D.M., Johnson, R., Reznek, R., 2004. Evaluation of the response to treatment of solid tumors — a consensus statement of the International Cancer Imaging Society. Br. J. Cancer 90, 2256–2260.
22
C.P.H. Vreuls et al. / Micron 84 (2016) 17–22
Jones, R.J., Lee, K.S., Beschorner, W.E., Vogel, V.G., Grochow, L.B., Braine, H.G., Vogelsang, G.B., et al., 1987. Venoocclusive disease of the liver following bone marrow transplantation. Transplantation 44, 778–783. Nagase, H., Visse, R., Murphy, G., 2006. Structure and function of matrix metalloproteinases and TIMPs. Cardiovasc. Res. 69, 562–573. Nakano, H., Oussoultzoglou, E., Rosso, E., Casnedi, S., Chenard-Neu, M.P., Dufour, P., Bachellier, P., et al., 2008. Sinusoidal injury increases morbidity after major hepatectomy in patients with colorectal liver metastases receiving preoperative chemotherapy. Ann. Surg. 247, 118–124. Narita, M., Oussoultzoglou, E., Fuchshuber, P., Pessaux, P., Chenard, M.P., Rosso, E., Nobili, C., et al., 2012. What is a safe future liver remnant size in patients undergoing major hepatectomy for colorectal liver metastases and treated by intensive preoperative chemotherapy? Ann. Surg. Oncol. (e-pub ahead of print). Overman, M.J., Maru, D.M., Charnsangavej, C., Loyer, E.M., Wang, H., Pathak, P., Eng, C., et al., 2010. Oxaliplatin-mediated increase in spleen size as a biomarker for the development of hepatic sinusoidal injury. J. Clin. Oncol. 28, 2549–2555. Robinson, S.M., Mann, J., Vasilaki, A., Mathers, J., Burt, A.D., Oakley, F., White, S.A., et al., 2013. Pathogenesis of folfox induced sinusoidal obstruction syndrome in a murine chemotherapy model. J. Hepatol. Rubbia-Brandt, L., 2010. Sinusoidal obstruction syndrome. Clin. Liver Dis. 14, 651–668. Rubbia-Brandt, L., Audard, V., Sartoretti, P., Roth, A.D., Brezault, C., Le Charpentier, M., Dousset, B., et al., 2004. Severe hepatic sinusoidal obstruction associated with oxaliplatin-based chemotherapy in patients with metastatic colorectal cancer. Ann. Oncol. 15, 460–466.
Rubbia-Brandt, L., Lauwers, G.Y., Wang, H., Majno, P.E., Tanabe, K., Zhu, A.X., Brezault, C., et al., 2010. Sinusoidal obstruction syndrome and nodular regenerative hyperplasia are frequent oxaliplatin-associated liver lesions and partially prevented by bevacizumab in patients with hepatic colorectal metastasis. Histopathology 56, 430–439. Soubrane, O., Brouquet, A., Zalinski, S., Terris, B., Brezault, C., Mallet, V., Goldwasser, F., et al., 2010. Predicting high grade lesions of sinusoidal obstruction syndrome related to oxaliplatin-based chemotherapy for colorectal liver metastases: correlation with post-hepatectomy outcome. Ann. Surg. 251, 454–460. van den Broek, M.A., Olde Damink, S.W., Driessen, A., Dejong, C.H., Bemelmans, M.H., 2009. Nodular regenerative hyperplasia secondary to neoadjuvant chemotherapy for colorectal liver metastases. Case Report Med. 2009, 457975. van den Broek, M.A., Vreuls, C.P., Winstanley, A., Jansen, R.L., van Bijnen, A.A., Dello, S.A., Bemelmans, M.H., et al., 2013. Hyaluronic acid as a marker of hepatic sinusoidal obstruction syndrome secondary to oxaliplatin-based chemotherapy in patients with colorectal liver metastases. Ann. Surg. Oncol. 20, 1462–1469. Vreuls, C., Wisse, E., Duimel, H., Stevens, K., Verheyen, F., Braet, F., Driessen, A., et al., 2013. Jet-fixation: a novel method to improve microscopy of human liver needle biopsies. Hepatology. Wisse, E., Braet, F., Duimel, H., Vreuls, C., Koek, G., Olde Damink, S.W., van den Broek, M.A., et al., 2010. Fixation methods for electron microscopy of human and other liver. World J. Gastroenterol. 16, 2851–2866.
