A 57-year-old woman presented with PET/CT scan evidence of a colonic mass with distention in addition to 12 intrahepatic FDG avid lesions, mainly confined to the superior/posterior aspect of the right lobe, but with some minor involvement of segments I and II. Colonic biopsy confirmed the presence of a poorly differentiated adenocarcinoma with focal signet ring features. She underwent colon resection and a wedge liver biopsy, which confirmed CRC hepatic metastases. After surgery for her primary tumor, she received FOLFOX and bevacizumab to facilitate resection of the remaining liver lesions. After partial response to preoperative therapy, she underwent right hepatectomy and wedge resection of segments I and II, with one R1 margin receiving additional cauterization. Despite postoperative completion of a total of 12 cycles of adjunctive chemotherapy, 15 months later CT revealed recurrent disease in her pelvis.
Case Discussion:
This is a 57-year-old woman underwent a colonic resection for a poorly differentiated adenocarcinoma with signet-ring features and a single hepatic wedge resection in August 2006. When you examine the overall metastasis rate for colon cancer in general, 60 percent of patients die of their disease. Approximately 35 percent present with synchronous primary tumor and metastases, while the other 25 percent have metachronous metastases.
After surgery, 10 of 14 lymph nodes were positive and 12 liver metastases were still visible on PET/CT scan, mainly confined to the right lobe. Postoperatively she received modified FOLFOX6 with bevacizumab.
The patient received six cycles with bevacizumab and then two more cycles of FOLFOX alone. She had a partial response to the therapy and was considered eligible for metastatectomy. She underwent a right hepatectomy and wedge resection of segments I and II. Ten metastases were identified in pathology while the other two metastases had apparently disappeared. Of course, it’s difficult at times to correlate the resected specimen with the imaging scans.
One segment had a positive surgical margin, but it was cauterized. I’ve discussed the impact of a positive margin on prognosis with Steve Curley and he says that in a R1 resection, cauterization alone provides so much additional margin that the positive margin does not impact the local recurrence risk. Personally, I’ve yet to see a local recurrence after a resection for colon cancer, whereas I have seen local recurrences after RFA.
During surgery, the patient also had a prior ileostomy taken down. Postoperatively, she developed septic complications from an anastomotic leak and had to be hospitalized. I spoke with a surgeon about this and he felt it was probably related to the bevacizumab, despite the fact that the agent had been omitted during the last two cycles of therapy and six weeks had definitely passed between the patient’s last dose and surgery.
The patient returned for postoperative therapy four or five months after surgery. We were outside the 12-week window we usually allow for adjuvant therapy, but the patient was clearly at high risk and anxious to receive chemotherapy, so we administered another four cycles.
A few months later, a questionable fistula with fluid collection was detected in the left upper quadrant in the area of the prior surgery. The patient was asymptomatic. We were unable to determine whether this was a bevacizumab-mediated gastrointestinal perforation or a complication of surgery. She seemed to have wound-healing problems, more or less. She had a number of comorbidities, including a seizure disorder, and her care was quite complicated.
Fifteen months after surgery, CT revealed large pelvic metastases. She had also developed several peritoneal metastases along the way, in addition to lymphadenopathy in the retroperitoneum, so we were now in a palliative situation.
This patient was first diagnosed with colon cancer in the pre-K-ras era. Had I seen her today, I would have tested the K-ras status and treated with cetuximab rather than bevacizumab, if it was the wild type, for a couple of reasons. First of all, the response-inducing capacity of cetuximab in K-ras wild type tumors is stronger than bevacizumab, particularly in combination with FOLFOX. The OPUS trial data showed a 24 percent increase in the response rate when cetuximab was combined with FOLFOX — that’s stronger than anything we’ve seen with bevacizumab.
Secondly, cetuximab is not associated with wound-healing problems, so it makes sense to use it in the perioperative setting. We now routinely test tumors for K-ras and B-ras mutations. These mutations are mutually exclusive — approximately 40 percent of cases will be K-ras mutant and another 10 percent will be B-ras mutant, so these tests will identify the 50 percent of patients who should not receive cetuximab.
However, our goal for this patient now is to gain time and maintain quality of life. In the palliative setting, bevacizumab is superior to cetuximab, even in patients with K-ras wild type tumors, based on cross-trial comparisons. Thus, we treated her with FOLFIRI/bevacizumab and she responded well to treatment. Her CEA level is almost normalized again.
