A 64-year-old woman presented with a potentially obstructing mass in her transverse colon after a routine colonoscopy. Abdominal CT showed evidence of three peripherally located (two centimeters or smaller) synchronous hepatic lesions in the right liver. She had a stent placed and received six cycles of preoperative FOLFOX. Six weeks later, she underwent resection of her primary tumor and surgical exploration of her liver metastases. An intraoperative ultrasound revealed only a 1.5-cm calcification in the liver which was ablated with RFA. Also, liver biopsy revealed mild macrovesicular steatosis. Postoperatively, she received adjuvant FOLFOX, at which time she was hospitalized for a severe allergic reaction to oxaliplatin and then continued on with 5-FU/leucovorin.
Case Discussion:
This is a 64-year-old woman who presented in March 2008 with a transverse colon cancer detected on a screening colonoscopy. It was Grade II out of IV. The pathologists here at the Mayo Clinic are some of the few who use a four-tier, rather than a three-tier, grading system for colon cancer. Grade IV of IV is significantly undifferentiated and Grades II and III are the moderately differentiated tumors.
Also at our institution, we always perform a staging CT scan before these patients go to surgery. Her CT revealed three hepatic lesions in the right side and in a subsequent PET scan, no extrahepatic disease was seen.
The patient was asymptomatic and the tumor was deemed potentially obstructing. A stent was placed and she was treated with modified FOLFOX6. I did not use a biologic agent because we wanted to test the chemosensitivity of her disease, and this was also in the pre-K-ras era. Today I probably would have added cetuximab if I had known the tumor was K-ras wild type.
As for bevacizumab, I was hesitant because the two perforations that I have seen with this agent occurred in patients with stents. I felt that this case did not require bevacizumab and the stent put her at high risk, so I didn’t use it.
We did not feel it was necessary to biopsy the hepatic lesions because they were PET-positive, the patient’s CEA level was elevated and she had a concurrent tumor in the transverse colon.
I don’t normally order a second PET scan before surgery because we commonly see false-negatives with patients on chemotherapy. The reason for that is a PET scan utilizes the functional capability of tumor cells to metabolize glucose. In response to chemotherapy, cells might shut down their metabolism and enter a dormant stage where they are not PET-positive, but they are still alive.
Also, I generally send patients like this to surgery within two or three weeks after preoperative therapy. However, I was away and this patient did not go to surgery until about six weeks after chemotherapy. She did have another PET and CT scan, which showed the colon mass was smaller and the hepatic lesions were no longer visible.
When the patient went to surgery for synchronous resection of the colon mass and the liver, three of the liver metastases were not detectable by palpation or ultrasound. A 1.5-centimeter cluster of calcifications was found in segment VII of the liver, which was interpreted as a remnant of a prior metastasis. This area was ablated, which was a good call, because it was quite deep and a resection would have sacrificed liver parenchyma.
The data show that if one can identify and resect the area where the lesions were, you will find viable tumor in 85 to 90 percent. In this case, the surgeon did not resect the liver because the other three tumors were scattered and not located superficially. The problem is that when you examine this large organ intraoperatively — that has some architectural changes secondary to the chemotherapy — it’s difficult to identify these areas, especially in a patient like this who had a complete response even by ultrasound. The choice then is to either resect the right lobe or to perform minimally invasive surgery, follow the patient and resect the lesions if they recur.
The surgeon did biopsy the liver and found some steatosis with a hepatitis component and portal lymphocytic infiltrate. I found this interesting, because steatohepatitis is more often linked to irinotecan-based therapy, but apparently it can happen with a FOLFOX-based regimen as well.
The tumor was tested and found to express wild-type K-ras. However, she had responded extremely well to prior chemotherapy, so we didn’t see any reason to add a biologic to her treatment postoperatively. After she recovered from surgery, she resumed FOLFOX, but unfortunately after the first cycle she developed a severe reaction to oxaliplatin and was briefly hospitalized.
In examination of the data from the NCCTG-N9741 trial, seven percent of patients receiving oxaliplatin had allergic reactions that led to treatment discontinuation. It doesn’t generally occur early on — rather, the median time to onset is approximately six cycles. When we see allergic reactions to this agent, we can either chose an alternative regimen or try to desensitize the patient. To do so, we use these miniscule amounts, diluted one to 100,000, on an in-patient basis and ramp up the infusion rate and the dose over time. The process takes about eight hours and there’s approximately a 50-percent chance that the desensitization will succeed.
We didn’t feel the need to desensitize this patient, nor did we consider FOLFIRI because we were in a postoperative setting and unable to evaluate response. When you examine the data comparing 5-FU versus FOLFIRI in the postoperative setting, no benefit is seen from the addition of irinotecan for patients who had no prior test for efficacy of irinotecan preoperatively. After her reaction to oxaliplatin, this patient continued to receive only the infusion 5-FU/leucovorin.
