Abstract
Background: Fluorouracil plus leucovorin (5-FU/LV) is a less toxic but mild chemotherapy. Case Report: A 63-year-old male patient with rectal cancer and multiple colorectal liver metastases (CRLM, total volume of 1,826 ml) was hospitalized. He had several poor prognostic factors, including elevated levels of tumor markers, with carcinoembryonic antigen and carbohydrate antigen 19-9 levels of 17,119 ng/ml and 7,617 U/ml, respectively. Additionally, the patient had a low body mass index, poor performance status, and a history of apparent weight loss. After capecitabine and oxaliplatin for four cycles, 5-FU/LV has been lasting for nine months. Interestingly, tumor marker levels returned close to normal limits, and the total CRLM volume decreased to 154 ml without any enhancements. The patient’s general condition clearly improved after a year of chemotherapy. Conclusion: Chemotherapy with 5-FU/LV and percutaneous microwave ablation is beneficial to achieve tumor control in patients with highly advanced liver-only CRLM and poor general condition.
Systemic chemotherapy with cytotoxic and targeted agents is the standard therapy for patients with unresectable colorectal liver metastases (CRLM) (1-3). Patients must be first evaluated as fit or unfit for chemotherapy based on their medical conditions. The type and intensity of induction chemotherapy are selected depending on the purpose of cytoreduction or disease control. Moreover, primary tumor location and RAS/BRAF status are essential to determine therapeutic drugs. Conversion hepatectomy is recommended for patients who had initially unresectable CRLM that changed to resectable CRLM via induction chemotherapy (4-6). Conversion surgery using thermal ablation is an attractive option for small CRLM under optimal control of chemotherapy (7-9). Thermal ablation includes radiofrequency ablation (RFA) and microwave ablation (MWA).
Currently, fluorouracil plus leucovorin (5-FU/LV) monotherapy is administered only to CRLM patients determined unfit for oxaliplatin, irinotecan, and targeted agents (1-3, 10). The regimen is also sometimes used during the maintenance period of standard chemotherapy. A previous meta-analysis study has demonstrated that the median survival time was 12.7 months for patients treated with 5-FU/LV, and the performance status (PS) was the only independent predictor of overall survival (11).
Pretreatment levels of tumor markers are correlated with tumor volume and tumor activity (12-14). Carcinoembryonic antigen (CEA) is a well-known tumor marker for colorectal cancer and a prognostic factor of unresectable CRLM. Patients with CEA levels >100 ng/ml are candidates for a poor prognosis. CEA levels over 15,000 at the time of diagnosis are rarely observed in nationwide Japanese CRLM cohorts, with only 0.2% in 3,820 patients between 2005 and 2007 and 0.3% in 3,525 patients between 2013 and 2014 (15, 16). Similarly, high carbohydrate antigen 19-9 (CA 19-9) levels (>100 U/ml) has been reported as an independent factor of poor prognosis in upfront and rescue hepatic resection (17, 18). CA 19-9 positive cancer cells tend to adhere to cultured endothelial cells activated by some inflammatory cytokines, developing hematogenous metastasis (19).
Numerous prognostic factors in chemotherapy for unresectable CRLM include tumor, genetic, patient, and treatment factors (1-3, 6). Response to chemotherapy is a strong prognostic factor and is usually assessed by tumor regression and tumor necrosis as well as a decrease in tumor marker levels (20-22). The normalization of tumor marker levels after chemotherapy is a significant predictor of good recurrence-free and overall survival followed by hepatectomy (23). The half-life of CEA levels can predict the therapeutic effects of induction chemotherapy for CRLM (24).
In this study, we present a patient with highly advanced CRLM with rectal cancer that was successfully controlled with continuous mild systemic chemotherapy in combination with MWA.
Case Report
A 63-year-old male visited the referral hospital because of general fatigue and weight loss. He was sent to our hospital with a diagnosis of advanced rectal cancer with bilateral CRLM. Lower gastrointestinal endoscopy revealed a circumferential type 3 lesion, 8 cm in length in the rectosigmoid colon. Computed tomography (CT) showed a rectal lesion with a thickened wall with some swollen regional lymph nodes and numerous CRLM (>10 cm in diameter and >30 nodules) (Figure 1A-D). The total CRLM volume calculated by the Synapse VINCENT system (25) was 1,826 ml. The stage of rectal cancer was classified as stage IV [T4b, N1, M1(H3)] as per the 7th revision of the AJCC/UICC TNM staging system. The biopsy of the rectal lesion revealed moderately differentiated adenocarcinoma with p53 over-expression and increased Ki-67 labeling index (91%), CEA expression, and CA 19-9 expression (Figure 2). Gene analysis showed mutant-type RAS and wild-type BRAF genes. The patient had a low BMI (22.5), poor PS of Eastern Cooperative Oncology Group (ECOG) grade 2, and apparent weight loss (7.2 kg per year). His skeletal mass index was 43.6 cm2/m2 and was not classified into the sarcopenia criteria (<43 cm2/m2) but showed slightly decreased visceral fat amount (94.8 cm2/<103 cm2/) (26, 27). The laboratory data on admission showed an extremely high CEA level of 17,119 ng/ml (≤5 ng/ml) and a CA 19-9 level of 7,617 U/ml (≤39 U/ml). The primary rectal lesion showed severe stenosis; thus, a metallic stent was placed. Capecitabine and oxaliplatin (Cape OX) were started three days after stent placement. Following four cycles of Cape OX, the patient had severe diarrhea (grade 3) as per the 5th version of Common Terminology Criteria for Adverse Events due to capecitabine. Additionally, massive ascites appeared possibly due to veno-occlusive disease caused by oxaliplatin. Thus, chemotherapy was discontinued. Ascites was well controlled with diuretic drugs and two times of cell-free and concentrated ascites reinfusion therapy. Chemotherapy with 5-FU/LV was started after 40 days of discontinuation, mainly due to mild adverse events. About 8 months after starting treatment, contrast-enhanced CT revealed two CRLM with early enhancement; therefore, percutaneous MWA was performed twice. The recent enhanced CT showed excessively diminished CRLM with calcification (Figure 1E-H). The total CRLM volume evidently decreased to 154 ml. The dynamic magnetic resonance imaging (MRI) showed no enhancement of CRLM. The treatment course and tumor marker levels are listed in Figure 3. Surprisingly, CEA and CA 19-9 levels returned close to the standard limit (8.5 ng/ml and 12.7 U/ml, respectively). The half-lives of CEA levels were 15 days and 66 days during chemotherapy with Cape Ox and 5-FU/LV, respectively. The patient showed improvement of general condition without any viable lesions after one year of chemotherapy.
Pre-chemotherapy abdominal computed tomography (CT) images of liver tumors (A-D). A plain CT image showed multiple hypodense lesions with fine calcifications in the liver. Eight months after chemotherapy, contrast-enhanced CT images from the portal venous phase showed heavy calcification occupying metastases without enhancement (E-F). Additionally, the low-density area made by the microwave ablation (arrow) can be detected.
Histological analysis of biopsy specimen. Hematoxylin and eosin staining and immunohistochemistry of p53 (clone DO-7, DAKO, Glostrup, Denmark), Ki-67 (clone MIB-1, DAKO), CEA (clone II-7, DAKO), and CA 19-9 (clone 1116-NS-19-9, DAKO) are presented. Non-immune mouse IgG was used as negative control.
The treatment course and changes in tumor marker levels. CEA: Carcinoembryonic antigen; CA 19-9: carbohydrate antigen 19-9; Cape OX: capecitabine and oxaliplatin; 5-FU/LV: fluorouracil + leucovorin; CART: cell-free and concentrated ascites reinfusion therapy; MWA: microwave ablation.
Discussion
Our patient showed liver-only metastases but had several poor clinical prognostic factors, including synchronous, large-sized, and multiple CRLM; lymph-node metastases of the primary lesion; and high levels of tumor markers (6, 14, 18, 21). Furthermore, positive p53 expression and high Ki-67 labeling index in the histochemical staining are independent unfavorable prognostic factors (28). CRLM patients with poor PS, low BMI, and preoperative weight loss are usually classified unfit for aggressive chemotherapy (1-3). One meta-analysis demonstrated that PS grade is the strongest predictor of survival in patients with metastatic colorectal cancer treated with 5-FU monotherapy and 5-FU/LV or 5-FU plus methotrexate (29). Median survival times were 14.6, 10.3, and 4.8 months for patients with ECOG grades 0, 1, and 2, respectively. A pooled analysis of five randomized trials of 5-FU-based chemotherapy showed that a low BMI (<25) was associated with a higher risk of hematologic toxicities (anemia and neutropenia) and poor overall survival patients with metastatic colorectal cancer compared with those with a high BMI (≥25) (30). Another study showed that >15% of weight loss six months before chemotherapy was significantly associated with short progression-free and overall survival (31). The current patient had a low BMI (22.5), poor PS (ECOG grade 2), and apparent weight loss (7.2 kg per year) at the start of chemotherapy.
Bevacizumab was not administered to our patient because of colorectal stenting. Standard Cape OX chemotherapy was discontinued after four cycles because of severe adverse events. After controlling diarrhea and massive ascites, 5-FU/LV monotherapy was started as palliative chemotherapy. The total volume of CRLM decreased from 1,826 ml to 429 ml after Cape OX therapy, which further decreased to 154 ml after 5-FU/LV monotherapy. Partial response according to the RECIST criteria has continued with no severe adverse events. Dynamic CT and MRI revealed that all CRLM had decreased in size and were accompanied by apparent calcification without any enhancement. Furthermore, no CRLM showed diffusion restriction in diffusion MRI. It was difficult to evaluate primary rectal cancer viability because of the metal stent, but no enlargement and enhancement was observed. Moreover, the exceptionally high pre-chemotherapeutic levels of CEA (17,119 ng/ml) and CA 19-9 (7,817 U/ml) were close to normalized by chemotherapy with 5-FU/LV followed by percutaneous MWA. We had reported that a shorter half-life (<20 days) is a predictor for good tumor response and pathological response (24). The half-life of our patient was 15 days through the Cape OX chemotherapy and 66 days through the 5 FU/LV. CRLM showed slight calcification pre-chemotherapy, and dense calcification developed with the increase in chemotherapy. Calcification of CRLM is detected among 12%-27% of patients with CRLM, and it is an independent good prognostic factor after chemotherapy (32). Tumor calcification during chemotherapy also predicts a high objective response rate and good overall survival (33).
For histopathological examination we only have a biopsy specimen of the primary colorectal tumor fixed to formalin. Therefore, it is difficult to resolve why a traditional 5-FU/LV is highly effective for the patient in our study. However, folypolyglutamated synthetase activity measured in CRLM tissue, which regulates thymidylate synthase, has been reported to be critical in assessing the responsiveness of 5-FU/LV (34). A previous study demonstrated that the protein expression of the 5-FU metabolic enzyme orotate phosphoribosyltransferase positively correlated with few adverse events and long patient survival of 5-FU-based chemotherapy (35). Furthermore, if resection of colorectal cancer and CRLM is possible, we may elucidate the reasons for the unexpected therapeutic effects.
The CLOCC study (EORTC 40004) was conducted as a randomized controlled phase II study comparing patients undergoing chemotherapy following RFA and those undergoing chemotherapy alone (36). Progression-free survival as a primary endpoint was significantly better in the chemotherapy plus RFA group compared with the chemotherapy alone group. Overall survival was comparable in the two groups; however, if the CRLM showed no progression during chemotherapy, overall survival of the chemotherapy plus RFA group was excellent and equivalent to that of the liver resection group in the EPOC trial (37). In our patient, the viable residual lesions were completely ablated with percutaneous MWA.
The observation period was too short to obtain conclusive results; however, we could show that the patient experienced excellent therapeutic effects despite multiple poor prognostic factors, including far-advanced CRLM status and poor general condition. Therefore, 5-FU/LV is still a promising option for selected patients with highly advanced but liver-only CRLM.
Footnotes
Authors’ Contributions
KY, KM, and TB identified the concept and wrote the draft of the article. YK investigated pathological findings. All Authors treated the patient and collected data. All Authors read and approved the final version of the manuscript.
Conflicts of Interest
The Authors have no conflicts of interest in relation to this study.
- Received December 31, 2021.
- Revision received January 17, 2022.
- Accepted January 18, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
