Impact of CAPOX or FOLFOX 4 on Spleen size , Platelet Count and Liver Function when Partnered Cetuximab as First-line Treatment for KRAS Wild-type Metastatic Colorectal Cancer

Objectives: Oxaliplatin can cause hepatic sinusoidal injury and splenomegaly. It remains unknown if the magnitude of injury would differ when oxaliplatin is combined with capecitabine or 5-FU with/without cetuximab. We investigated the impact of 1 line CAPOX or FOLFOX4 and the additional cetuximab on spleen size, platelet count and liver function in patients with KRAS wild-type metastatic colorectal cancer (mCRC). Methods: 101 Patients planned to receive either CAPOX or FOLFOX4 with/without cetuximab as first-line treatment were prospectively recruited. Changes in spleen size by volumetric measurement after treatment were determined. Correlation studies were performed for factors associated with changes in spleen size, thrombocytopenia and impaired liver function. Results: The spleen enlarged (median +17.9%, P < 0.001) after treatment. Multivariable analysis revealed that capecitabine, its dose intensity and cumulative dose (per 10000mg increase) correlated with splenomegaly (P = 0.01, P = 0.02 and P = 0.006, respectively). Increase in spleen size (P = 0.004) and splenomegaly (P = 0.002) correlated with thrombocytopenia. Dose intensity and cumulative dose of capecitabine (per 10000mg increase) and increase in spleen size correlated with grade 1 impaired liver function (P = 0.01, P = 0.003 and P = 0.04, respectively). Use of cetuximab correlated with less splenic enlargement (+13.7% vs. +22.7%; P = 0.04), especially when coupled with FOLFOX4 rather than CAPOX (+1.1% vs. + 23.0%; P = 0.003). Conclusions: Capecitabine was associated with more splenomegaly which in turn correlated with thrombocytopenia and impaired liver function. Cetuximab offered some protection from further splenic enlargement especially when combined with FOLFOX4.


77
Fluoropyrimidines and oxaliplatin have been used in metastatic colorectal cancer (mCRC) 78 for more than ten years. 1  between oxaliplatin plus infusional 5-FU and oxaliplatin plus capecitabine was not 88 randomized but it was an agreement between patients and treating physicians before 89 treatment commencement. Its first publication reporting its toxicities profiles revealed that 90 severe grade 3/4 diarrhea was observed in 30% of patients who received oxaliplatin and 91 capecitabine, leading to study protocol amendment in 2007 with dose reduction of 92 capecitabine from 1000mg/m 2 to 850mg/m 2 twice daily in future patients. 12 This may be 93 one of the reasons of failure to improve overall survival as published in 2011. 13 At the same 94 time, with growing experience of using oxaliplatin in the past decade, this drug was also 95 6 noted to have close association with hepatic sinusoidal injury and post-hepatectomy 96 morbidity and mortality when given pre-operatively. [14][15][16][17][18][19] Moreover increase in spleen size 97 was recently proven an effective biomarker for such hepatic adverse event after 98 oxaliplatin. 20 This adverse hepatic complication is definitely a particular concern to the 99 surgeons and patients when perioperative chemotherapy is increasingly adopted for 100 potentially resectable liver metastases. 21

116
Hilden, Germany), followed by polymerase chain reaction (PCR) amplification and direct 117 sequencing using enriched tumour genomic DNA before treatment. 24 All patients had 118 baseline contrast-enhanced computed tomography (CT) scans of 5mm slice thickness of the 119 thorax, abdomen and pelvis performed for the confirmation of primary tumour if present, 120 regional nodal involvement and distant metastasis by board-certified radiologists within 4 121 weeks before treatment. After baseline physical examination and blood tests for hematology 122 and biochemistry, they discussed with their treating physicians for choices between CAPOX 123 and FOLFOX4 and whether to add cetuximab in addition to the chemotherapy regimen as a 124 self-financed drug. Baseline liver impairment was defined according to liver biochemistry.

128
These values were then summed up to give a baseline liver impairment score, which was 129 further subclassified into 4 groups (baseline liver impairment score 0, 1-4, 5-8 and 9) 130 accordingly, slightly modified and adapted from the criteria devised by Twelves et al. 26 131 CAPOX was given as a 3-weekly regimen with oxaliplatin 130mg/m 2 infused over 2 hours 132 on day 1 followed by capecitabine 1000mg/m 2 orally twice a day for 2 weeks followed by a

217
Grade 1 and grade 3 impaired liver function was noted in 51.5% and 1.0% respectively.

231
An example from one patient whose spleen enlarged after CAPOX plus cetuximab was 232 illustrated in Fig. 2. Splenomegaly was also significantly associated with thrombocytopenia 233 (P = 0.02) and marginally associated with grade 1 impaired liver function (P = 0.05).  (Table 3).

333
KRAS wild-type mCRC had been well established in CRYSTAL and OPUS study. In our 334 study, patients' decision on either FOLFOX4 or CAPOX was mainly based on their concern 335 about financial affordability, hospitalization and their own preference. The decision of 336 adding cetuximab or not was purely their financial consideration as they had to pay at their 337 own cost for cetuximab. In fact, no uneven distribution of the presence, number and volume 338 of liver metastases was found in our patients who received FOLFOX plus cetuximab and 339 CAPOX plus cetuximab (P = 0.24, P = 0.91 and P = 0.17 respectively). Most importantly, 340 unvariable and multivariable analyses did not reveal these factors as predictors of our four 341 treatment outcomes. 19

343
In conclusion, our study demonstrated that CAPOX should not be the preferred