br Results br Patient Characteristics br Data from patients
Data from 474 patients were included in the database; of these 460 were valid data entries. Cases on the register with no diagnosis or no SIRT procedure date were excluded, leaving 399 valid CRC patients for the analysis; patients were followed up for a median of 14.3 months (95% confi-dence interval 9.2e19.4). Fifteen cases were excluded for the following reasons: (i) missing diagnosis; (ii) no treat-ment data; (iii) no SIRT procedure date. Sixty-seven per cent of patients were men and had a median age of 66 years. Most patients had an ECOG performance score of 0 or 1 (93%) and most did not have extrahepatic metastatic dis-ease (60%) (Table 1). Almost all patients (98%) had received prior systemic chemotherapy or biologics and 78% had received two or three lines of prior chemotherapy, con-sisting predominantly of fluoropyrimidine-, oxaliplatin- or irinotecan-based regimens. The median duration from pri-mary and metastatic diagnoses to the first SIRT procedure was 2.1 years and 1.8 years, respectively (Table 1).
Treatment and Follow-up
Relevant visceral Pladienolide B were embolised during the work-up procedure in 52% of patients (Table 2). As is the practice in the UK, most patients received SIRT as a single procedure targeting the whole liver (52% split microsphere administration; 17% single microsphere administration). A very small proportion (3%) of patients had sequential lobes treated in two (or more) sessions. Based on the prior expe-rience of the treating centres, most SIRT treatments (86%) were conducted using resin Y-90 microspheres, with a mean prescribed activity of 1.74 GBq (standard deviation 2.13); 14% of treatments used glass Y-90 microspheres, with a mean prescribed activity of 4.18 GBq (standard deviation 1.71). Most patients (88%) had a hospital stay of 1 or 2 nights for the treatment. Chemotherapy was delivered concomitantly in 35% of cases (predominantly 5-fluorouracil and oxaliplatin) and a minority of cases (22%) went on to receive further post-SIRT chemotherapy during their follow-up phase (Table 2).
Table 1 Patient characteristics
Parameter Data for 399 patients
Total number of patients 399
Baseline ECOG score
Limited extrahepatic disease
Location of metastatic disease
Lymph nodes 40
Primary tumour resected
Number of previous
Prior chemotherapy received
Fluoropyrimidine-based 282 (71%)
No chemotherapy recorded 53 (13%)
Prior adjuvant therapy
Prior hepatic procedures*
ECOG, Eastern Cooperative Oncology Group; SIRT, selective internal radiation therapy; IQR, interquartile range.
* Hepatic surgical, ablative, vascular or radiotherapy procedures.
Subgroup analysis identified four covariates that resulted in a statistically significant difference in median overall survival (Table 3). Overall survival was significantly longer in patients who did not have extrahepatic metastasis (Log-rank test, P ¼ 0.021); the hazard ratio was 0.74 (95% confi-dence interval 0.57e0.96; univariate Cox proportional hazards P ¼ 0.022). Overall survival also differed signifi-cantly between the categories of number of liver tumours (Log-rank test, P ¼ 0.008); the hazard ratio was 1.67 (95% confidence interval 1.06e2.62; P ¼ 0.027) when the group of six to 10 tumours was compared with the reference group of one to five tumours; the hazard ratio was 1.61 (95% confidence interval 1.17e2.21) when the group of >10 tu-mours was compared with the reference group. Overall survival was longer in males compared with females (Log-rank test, P ¼ 0.012); the hazard ratio was 1.389 (95% con-fidence interval 1.073e1.800; P ¼ 0.013). Overall survival was also related to the percentage tumour to liver volume measurements at baseline (Log-rank test, P < 0.001); the hazard ratio was 1.955 (95% confidence interval 1.424e2.685) comparing the category of tumour to liver volume >25% to 50% with the reference category of ¼25%; the hazard ratio was 2.994 (95% confidence interval 1.791e5.005) when the category of >50% was compared with the reference category. No significant difference in survival was observed using the covariates of prior chemotherapy lines, ECOG performance status, age and prior liver procedures (Table 3).
Progression or death was observed in a total of 331 (269 patients’ disease progressed [67%]; 62 patients died before progression [16%]) and 24 (6%) patients were censored at the last imaging date when no progression was recorded. The progression status of 24 patients (6%) was unknown. The median PFS was 3.0 months (95% confidence interval 2.8e3.1) (Figure 2). Liver-specific progression or death was observed in 299 (75%) patients, 53 (13%) patients were censored and 43 (11%) were excluded. The median LPFS was 3.7 months (95% confidence interval 3.2e4.3) (Figure 2). Hepatic progression and extrahepatic progression were recorded on the same date in 81% of patients where both dates were recorded. Extrahepatic progression occurred before hepatic progression in 16% of patients.