Identification and Management of post SIRT toxicities and complications

Transcription

1 Identification and Management of post SIRT toxicities and complications A/Prof. Lourens Bester Interventional Radiologist University of Notre Dame Sydney

2 Disclosure Consultant and Proctor for SIRTEX Medical Australia and Asia.

3 Objectives: 1. Injury to the liver (REILD) Identify the factors responsible for REILD and how to manage REILD. 2. Injury to other organs ( Non-targeted delivery) Identify the causes for non-targeted delivery, how to prevent it and how to manage it.

4 REILD

5 REILD Definition: Radioembolisation (RE)-induced liver disease HEPATOLOGY, Vol. 57, No. 3, 2013 (REILD) has been defined as jaundice and ascites appearing 1 to 2 months after RE in the absence of tumour progression or bile duct occlusion.

6 REILD Rare complication of SIRT = 2% Characterised by a constellation of findings: 1. Typical temporal relationship to SIRT. 2. Typical clinical picture. 3. Typical biochemical picture. 4. Typical histopathological picture.

7 REILD 1. Temporal relationship to SIRT - REILD typically manifests approximately 4 8 weeks post-sirt. 2. Typical clinical picture or REILD - REILD is characterised by jaundice and ascites in the absence of tumour progression or bile duct obstruction.

8 REILD 3. Typical biochemical picture: - elevated bilirubin (> 3mg/dL) (> 50µmol / L) in almost all cases. - elevated alkaline phosphatase (ALP) and GGT in most cases. - accompanied by virtually no change in the transaminases (AST and ALT) 4. Typical histopathological picture: - sinusoidal obstruction * Important to remember that some of the chemotherapeutic agents used also lead to sinusoidal obstruction / venous occlusive changes.

9 Sinusoidal obstruction. Sinusoidal obstruction / veno-occlusive disease of the liver refers to a form of toxic liver injury characterized clinically by the development of hepatomegaly, ascites, and jaundice, and histologically by diffuse damage in the centrilobular zone of the liver. The cardinal histologic features of this injury are marked sinusoidal fibrosis, necrosis of pericentral hepatocytes, and narrowing and eventual fibrosis of central veins and obstruction of liver blood flow. Oxaliplatin and the SIRFLOX Study.

10 The SIRFLOX Study To assess the efficacy and safety of adding targeted radiation (SIR-Spheres microspheres) to standard-of-care systemic chemotherapy (FOLFOX6m + bevacizumab), compared to FOLFOX6m chemotherapy (+ bevacizumab) alone as 1 st -line therapy in patients with non-resectable colorectal liver metastases, with or without evidence of extra-hepatic metastases Design: Prospective open-label, multi-centre, multi-national RCT Eligible Patients: Unresectable liver-only or liver-predominant metastatic CRC No prior chemotherapy for mcrc Fit for combination therapy and SIRT Stratify: Presence of extrahepatic metastases Degree of liver involvement Institution Use of bevacizumab Randomise 1:1 n = 518 SIR-Spheres FOLFOX6m * +bevacizumab C4 SIRT day 3 4, Cycle 1 = oxaliplatin 60 mg/m = Cycles 1 3 in chemo-sirt arm; Bevacizumab from Cycle 4 in test arm, Cycle 1 (or per institutional protocol) in control arm; FOLFOX6m + bevacizumab C1 Primary endpoint: Progression-free survival (PFS) Secondary endpoints: PFS in liver Overall survival Response rate Quality of life Recurrence rate Toxicity Resection rate

11 Risk factors of REILD REILD may occur: Non-cirrhotic and Cirrhotic patients. 1. In non-cirrhotic patients, the main risk factors for the development of REILD include prior exposure to systemic chemotherapy, extensive tumour infiltration and whole-liver SIRT. 2. In cirrhotic patients, the main risk factors are small liver volume (<1.5 L), whole-liver SIRT, and elevated bilirubin (>21 µmol / L) (1.2 mg/dl) at baseline. In the setting of first-line deployment of SIRT in a healthy liver like in SIRFLOX REILD is rare. REILD is largely confined to patients with a well-known liver condition such as cirrhosis, or those who have been exposed to any form of potential liver insult, most usually prior chemotherapy. HEPATOLOGY 2013;57:

12 J Cancer Res Clin Oncol 2013 Review 20 papers ( 979 Patients) between All patients failed 3 lines of chemotherapy. Median survival 12 months. Most cases of acute toxicity were mild (Grade I or II) resolved without intervention. Grade 4 less than 3%. The most common acute toxicities were fatigue (38.5 %), abdominal pain (16 %) and nausea/vomiting (19 %). Prognostic factors identified with a poor outcome: - The presence of extra-hepatic disease - Extensive pre-treatment chemotherapy ( 3 lines) - Extensive liver disease ( 26 %).

13 Eur J Nucl Med Mol Imaging 2014 The aim of this study was to analyse treatment characteristics effecting survival outcomes. Overall survival was analysed from: date of diagnosis of primary cancer, hepatic metastases and from the 1 st Yttrium 90. The most common acute toxicities were fatigue,abdominal pain,and nausea/vomiting. Median overall survival was 43.0, 34.6, and 10.6 months.. In a Multivariate analysis : Survival was significantly longer in patients: (1) who received 2 cytotoxic drugs (n=104) than those who received 3 (n=110) (15.2 vs.7.5 mo)( p=0.0001) (2) who received no biologic agents (n=52) compared with those that did (n=162) (18.6 vs. 9.4 months) (p=0.0001).

14 Median survival after Yttrium 90 Radioembolization was 10.5 months with 21% survival at 24 months. ORR: 72% Four factors were associated with poor prognosis: 1. Extensive tumour volume 2. Number of previous lines of chemotherapy 3. Poor radiological response to treatment 4. Low pre-op Haemoglobin. Clinical toxicity after treatment were minor (grade I/II) and resolved without active intervention. Conclusion: Yttrium 90 Radioembolization is a safe and effective treatment for unresectable, chemorefractory m-crc. Treatment at an earlier stage before chemoresistance develop and before extensive infiltration is present should be considered. Saxena, Bester et al Annals of Surgical Oncology. 2014

15 ECOG > 0 P= Sangro et.al Hepatology 2011;57: Kennedy et.al.int J Radiation Onc Biol Phys 2007; 68:1:13-23 Ibrahim et al W J Gastro 2008;21: Kennedy et al. Int.Cong.on Anti-Cancer Therapy (ICACT) 2008 Abs Salem et al. J Vasc Interv Radiol 2006; 17: Poor Prognosticators. Tumour volume of less than 10% or more than 70% P= An Infiltrative tumour presentation P= Infiltrative tumour pattern + low Albumin 30g/l p=0.02 AST & ALT more than 5 x normal p = 0.03 Bilirubin > 35 µmol/l P=0.0014

16 REILD Ascites, Jaundice, PHT and Diffuse Parenchymal Changes 8W 12W Fat Infiltration + Ascites

17 Capsular Retraction due to Hepatic Fibrosis and Portal Hypertension. Recovered Progression Atassi et al, RadioGraphics 2008; 28: 81-99

18 Prevention of REILD Activity calculation to minimize REILD while delivering tumoricial radiation This protocol minimizes REILD in HCC patients Important in those with poor liver function, keep D to maximum of 40 Gy Gil-Alzugaray et al Hepatology 2013;57:

19 Dosimetry BSA Model: Recommended formula for most patients Dose (GBq) = BSA (m 2 ) Partition Model: Assumption is for safety of liver exposure not tumor exposure Dose (Gy) = activity (GBq) x mass (g) volume tumour volume tumour + liver Partition model will allow predictive dosimetry.

20 Courtesy Andrew Kennedy

21 Prevention of REILD Treatment Plan. 1 Whole Liver 1 Single session: full dose in single session 2 2 Lobar Administration to one lobe of the liver 3 Segmental/Segmentectomy Selection into segment Segmentectomy: high dose radiation to whole segment 3 David M Liu MD

22 Management of REILD Three steps to be consider. 1. Steroids (starting from 0.5 mg/kg daily of methylprednisolone and rapid tapering) + ursodeoxicolic acid (600 mg/day). Low molecular weight heparin (fraxiparine 0.3mg 2 x/d for 8 weeks) could be added. Both steroids and heparin would only be useful if started very early in the course of the disease. 2. Defibrotide (starting from 2.5 mg/kg qid -10 mg/kg/dayand escalating every 3rd day 10 mg/kg/day if no response up to a maximum dose of 60 mg/kg/day or toxicity). It should be administered as a 2-hour infusion. The earlier it is started, the better it may work.

23 Management of REILD 3. TIPS (trans jugular intra-hepatic porto-systemic stent shunt placement). Hepatology recommendation is not to wait until synthetic liver function starts declining or bilirubin goes higher than µmol/l (10-15 mg/dl). If there is no "room" for placing TIPS in tumour-spared liver it may still be worth considering TIPS as dissemination of tumour cells could be better than immediate liver failure. 4. Prophylactic steroid use straight after implantation 0.5 mg/kg daily of methyl-prednisolone and rapid tapering.

24 Non- targeted delivery of SIRT

25 Radiation Pneumonitis Excessive deposition of radiation to the lungs, termed radiation pneumonitis include onset of dry cough, progressive exertional dyspnea 1-6 months after therapy, chest radiography and CT demonstrating excessive patchy consolidation with sparing of the lateral edges of the lungs and fissures. Identify with pre-treatment Tc 99m -MAA lung shunt study that demonstrates a potential for >30Gy exposure to the lungs. Eur J Nucl Med Mol Imaging August 2010

26 Reduction of Arterio-venous Lung Shunting Bester L, Salem R, J Vasc. Interv. Radiol. 2007;18:

27 Radiation Pneumonitis A safer methodology is a determination of exposure to the lungs based on the partition model. Utilising this model as opposed to the arbitrary dose reduction model currently followed will allow treating patients that are above 20% shunt level without the risk of pneumonitis. Example: 28% lung shunting on MAA but dose to the lungs is 24 Gy. ( Should preferably be below 30 Gy.) you can go ahead and treat the patient. Treatment is aimed at decreasing the inflammation. Steroids, such as prednisone, are given until the inflammation subsides and then slowly decreased over time.

28 Non Targeted Flow to the GIT and Other Organs. Balancing the appropriateness of the point of implantation with the potential of delivering particles in non-targeted areas remains the challenge of Radioembolisation. If the pre-treatment hepatic angiogram demonstrating a potential for deposition of microspheres in the GI tract or other organs that cannot be corrected by angiographic embolisation you can not proceed to do SIRT. The safety profile of SIRT ultimately depends on the identification and ability to understand the vascular anatomy of the liver and its vascular variants. The potential of nontargeted embolisation as a result of neovascularisation developing from previous optimized vascular beds prior to implantation has been reported further emphasising the need for identification of vascular supply up to the moment of implantation

29 MAA Free Tc 99m Pertechnetate Thyroid Thyroid

30 CT Arteriogram Use CT Angiography to determine non targeted flow. Tip of Catheter

31 Consequences of Non-Targeted flow. The most common complication as result of nontargeted embolisation is the deposition of microspheres within gastric circulation. The right gastric artery remains the most likely culprit, resulting in irradiation of the lesser curvature of the stomach with radiation ulcers that may range from self limiting to perforation resulting in surgery. Patient can be placed on prophylactic anti-ulcer medications e.g. PPI, H2 antagonists on day of SIRT and for 1-2 months post SIRT. Ulcerations do not necessary manifest acutely. It could be a sub-acute/chronic onset and it may manifest late 3-6 months post SIRT. QoL.

32 Consequences of Non-Targeted flow Cholecystitis is rare with less than 1% requiring surgical intervention. Pancreatitis Although described anecdotally, no specific case report relating to radiation-induced pancreatitis has been published in contemporary literature. Small bowel perforation / Radiation enteritis Pericarditis. Oesophageal ulceration. Radiation dermatitis.

33 Workup Angiogram Angiograms to be performed meticulously as described in the literature. Identify accessory vessels that can lead to non targeted flow. A good study reduces complication rates drastically. Use power injections, flood the arterial system with contrast and continue to acquire images till you see the parenchymal phase.

34 Variant Arteries. Watch out for the RGA usually from LHA If you can not coil think of reperfusion. RGA Watch out for the Oesophageal or Pericardial arteries Oesophageal

35 Reperfusion Technique. Oesophagus MHA J Vasc Interv Radiol 2011; 22:

36 Phrenic artery Use bland embolisation to occlude additional vascular tributaries where you would not like to use SIRT. With permission Dr Lam Vietnam Brugmans et al. J Vasc Interv Radiol Oct;23(10):

37 Artery of Falciform Ligament Do Nothing / Ice MAA Bester et al Clinical Nuclear Medicine 2008; 33(10):1-4 Dong et al AJR 1999;172: Lui et al VIR1999;10: Williams et al Radiology 1985;156:

38 Cystic Artery Do not have to coil the Cystic artery, the chance of developing Cholecystites is very low 0.6 2% Bester et al European Journal of Radiology Extra 2009; 72: e73 e74

39 Lost Coil You lost a coil! Leave it do not retrieve it, retrieval will damage the artery, the artery will not occlude, if the artery occlude it will reperfuse.

40 List of commonest complications incidence and management Lui et al.j Nucl Med Radiat Ther 2011; 2:

41 Chemotherapy

42 Tumour Vascularity and Avastin

43 Changes in the liver post Chemotherapy. Hepatic Fibrosis Biliary stasis Veno occlusive changes

44 Treatment

45 How do we manage backflow: Use 5% Dextrose instead of Water for Injection Vasodilators Slow injections 20ml syringes instead of 5ml syringes reduce pressure. Sandwich technique. Substitute WFI with half strength contrast: Conclusion: A greater percentage of the radioembolization dose was administered when using dilute contrast as the delivery medium. Termination for stasis occurred less frequently than when sterile water. Continuous real time monitoring mitigating the chance of reflux and non-target embolization. SIR Scientific Session 2014 J Vasc Interv Radiol 2003;

46 Thank you