Abstract
Background: Effectiveness and toxicity of transcatheter arterial injection of irinotecan-eluting beads (DEBIRI) with and without concurrent capecitabine in pre-treated patients with metastatic colorectal cancer (CRC). Patients and Methods: An Institutional Review Board-approved, multi-institutional registry from 5/2008 to 8/2013 was reviewed. Patients who received DEBIRI with (X-DEBIRI) or without (DEBIRI) capecitabine were compared. Results: Twenty-two X-DEBIRI and 149 DEBIRI patients were compared. There was no difference in the two groups with regards to adverse events (p=0.56). During a 3- and 6-month evaluation, the disease control rate (DCR) was similar in both groups. During the 12-month evaluation, there was better DCR in the X-DEBIRI group (p=0.03). Median survival was 13 months in the DEBIRI group and 22 months in the X-DEBIRI group (log-rank test, p=0.217). Conclusion: There is no additional toxicity when adding capecitabine with DEBIRI. Concurrent capecitabine may offer more durable disease control rate compared to DEBIRI-alone. Survival benefit with concurrent capecitabine was not statistically significant but there may be a trend towards improved survival.
- Irinotecan
- chemoembolization
- capecitabine
- drug-eluting-beads
- toxicity
Colorectal cancer (CRC) is one of the most prevalent malignancies in the world and the 3rd most common malignancy in the United States (24). The liver is the most common site of metastatic spread and the treatment of these hepatic metastases, rather than that of the primary neoplasm, usually dictates the patient's overall survival in patients with liver-dominant disease (8). Surgery in combination with chemotherapy provides the only chance of long-term 5-year survival but for patients who are not surgical candidates the overall survival is reduced. Fluorouracil (FU)-based therapy, irinotecan, oxaliplatin and biological agents (bevacizumab, regorafenib, etc.) are the mainstay in the treatment of this patients' subset. However, long-term tolerance (defined as >4 months) with these triplets and doublets are limited leaving few options for disease control maintenance (5, 12, 14, 21). The recent OPTIMOX-2 trial demonstrated that discontinuation of chemotherapy agents was associated with a negative impact on the duration of disease control and progression-free survival compared maintenance chemotherapy strategy (3). Recent data reported in 2014, 10th annual Gastrointestinal Cancers Symposium on January 26th in San Francisco, CA, found that bevacizumab-plus-capecitabine prolonged progression-free survival (PFS) in patients with metastatic colorectal cancer (mCRC) after receiving optimal induction chemotherapy.
Capecitabine (Xeloda™) is a potent oral chemotherapeutic agent that is recommended for adjuvant therapy and first-line therapy in patients with colorectal cancer (2, 20, 25). Unfortunately, its use in CRC patients, who are refractory to treatment, is questioned due to mixed reports on its efficacy and survival benefit (5, 11, 13, 23). However, it appears that Xeloda in combination provides better disease stabilization than when used alone.
The liver still remains a sanctuary area for disease control with chemotherapy-alone in patients who have progressed after first- or second-line therapy or after initial stabilization with induction therapy. We have recently reported on the effectiveness of hepatic arterial therapy with Irinotecan Drug Eluting Beads (DEBIRI) in patients with metastatic chemorefractory disease (18). DEBIRI has been shown to provide good efficacy in controlling the hepatic tumor burden and survival benefit has been suggested (10, 18, 22). Intrahepatic cancer is preferentially perfused by the hepatic arterial blood flow, whereas normal hepatic parenchyma is preferentially perfused by the portal system (1). The beads obstruct blood flow in the distal arterioles and capillaries, thereby increasing the dwell time and first pass extraction of the chemotherapeutic agent (6). It ultimately leads to high intra-hepatic tumor concentrations while limiting systemic exposure (6, 7, 16). While it provides potent hepatic tumor control, extrahepatic tumor control remains in jeopardy, hence the interest in evaluating the combination of systemic chemotherapy with DEBIRI in patients with metastatic colorectal cancer. Theoretically, if we can safely intensify chemotherapy exposure to the liver, which is the dominant site of disease, and also address extrahepatic disease with systemic chemotherapy, we hypothesize that this can improve overall outcome. We also wanted to investigate the possibility of synergy between the two treatments given that capecitabine is preferentially converted to 5-fluorouracil (5-FU) in tumor tissues.
To our knowledge, there is no study presenting the combination of capacitabine and DEBIRI in patients who have failed either first- or second-line chemotherapy. In the present study, we compared pre-treated patients treated with DEBIRI and concurrent capecitabine with patients treated with DEBIRI alone. We were interested in evaluating the toxicity, efficacy and survival of combination capacitabine and DEBIRI to assess whether this is a possible treatment option for patients with 2nd- and 3rd-line metastatic CRC.
Materials and Methods
Entry criteria. An Institutional Review Board-approved, prospectively maintained, multi-institutional, open, noncontrolled, repeat-treatment registry was evaluated from 5/2008 to 8/2013. A total of 297 patients presenting with liver dominant metastatic CRC were treated with DEBIRI.
Patients were included if they were 18 years old, of any race or sex, who had histologic or radiologic proof of metastatic disease to the liver, were able to give informed consent and eligible for treatment. The following were also required: Eastern Cooperative Oncology Group performance status score ≤2, life expectancy >3 months and lack of pregnancy with ongoing use of an acceptable contraceptive (in premenopausal women). Exclusion to bead therapy included contraindication to angiographic and selective visceral catheterization, significant extrahepatic disease representing an imminent life-threatening outcome, <75% hepatic parenchymal involvement, severe liver dysfunction, pregnancy and severe cardiac comorbidities. Only patients with liver-dominant disease (defined as >50% of overall total body disease burden) were considered for treatment. All chemo-naïve patients were excluded. All patients had disease progression with at least one previous line of chemotherapy (Table I).
Standard pre-therapy evaluation of patients with metastatic disease included, at minimum, three-phase computed tomography (CT) of the abdomen and pelvis and chest radiograph ≤1 month before treatment, including positron-emission tomography scanning depending on the institution and availability of the technology. Using the selection criteria, we identified 149 of 297 patients who were treated with DEBIRI without concurrent chemotherapy (DEBIRI) and 22 patients who were treated with DEBIRI and concomitant capecitabine (X-DEBIRI).
DEBIRI technique and capecitabine dosing schedule. Our Institutional technique for DEBIRI administration has been published elsewhere (17-19). Capecitabine was taken for 14 days, followed by a 7-day rest period for a total cycle time of 21 days. The dose of capecitabine was 1,250 mg/m2 administered orally twice daily (morning and evening; equivalent to 2,500 mg/m2 total daily dose).
Safety and efficacy variables and study schedule. Patients were followed-up for any treatment-related adverse experiences for 30 days after each treatment. All adverse events were recorded per standards and terminology set forth by the Cancer Therapy Evaluation Program's Common Terminology Criteria for Adverse Events (CTCAE) version 3.0. Data were also censored at the 30 day follow-up visit so that all adverse events occurring within 30 days of treatment were recorded and evaluated. Standard follow-up protocol included triphase computed tomography (CT) scan of the liver within 1-2 months post-treatment. Evaluation for tumor response was performed according to modified Response Evaluation Criteria in Solid Tumors (RECIST) criteria (4, 9, 15). Follow-up was then repeated every 3 months for the first year and every 6 months for the second year.
Overall survival (OS) is defined as the time between the treatment start date and death from any cause. Progression-free survival (PFS) is defined as the time between the start of treatment and objective disease progression or death. Hepatic-specific progression-free survival (HPFS) is defined as the time between the start of treatment and disease progression in the liver or death.
Statistical analysis. The Student's t and Fisher exact tests were used to compare continuous and categorical variables between the two groups, respectively. Survival and progression-free survival probabilities were generated with the Kaplan-Meier technique. Differences in the probability curves were assessed using the log-rank test. Multivariable Cox regression (backward stepwise) were used to evaluate the association between variables and survival. p-Values less than 0.05 were considered statistically significant. All statistics were calculated using the MedCalc software (version 12.7.8; MedCalc, Ostend, Belgium).
Results
Patients' characteristics. Patients' characteristics and pre-treatment chemotherapeutic agents are summarized in Table I. The median ages were 60 (range=51-86) and 64 (range= 52-82) for the X-DEBIRI and DEBIRI groups, respectively. A total of 52 procedures in 22 patients were performed in the X-DEBIRI group and 339 in 149 patients in the DEBIRI group. Technical success, defined as the ability to access the intended vessel and deliver at least 50% of the planned DEBIRI dose, was achieved in all patients. The median delivered dose was 92 mg for the X-DEBIRI group and 100 mg for the DEBIRI group (Table II).
Statistical analysis performed comparing the X-DEBIRI cohort group to the DEBIRI cohort group showed no difference in patient variables including extra-hepatic disease, prior surgery/ablation and hepatic tumor burden (Table I). Multivariate Cox regression identified that presence of extrahepatic metastatic disease (hazard ratio, 1.7; 95% confidence interval (CI), 1.06 to 2.7; p=0.03) and the total size of the target (hazard ratio, 1.06; 95% CI, 1.03 to 1.10; p≤0.001) were independently associated with decreased survival. The age, sex, extent of liver involvement, prior surgery, prior chemotherapy and number of liver lesions were not associated. Also, heterogeneous treatment factors, such as the use of doxorubicin-loaded drug eluting beads (DEBDOX) and radioembolization were not independent prognostic factors.
Adverse events. Adverse events are summarized in Table III. Overall, there was no statistical difference in the quantity and severity of all grade adverse events in the two groups (p=0.56). In the DEBIRI patient cohort, in a total of 339 treatments, there were 60 adverse events in 43 patients. In the X-DEBIRI group, after 52 total treatments, there were 11 adverse events in 2 patients. Furthermore, the frequency of high-grade adverse events (grade≥3) was not statistically different in both groups (p=0.49). The most common adverse events were pain in the DEBIRI group and nausea in the X-DEBIRI group. The single-grade three-event in the X-DEBIRI group was sepsis (one event in one patient). The high-grade events (17 events in 15 patients) encountered in the DEBIRI group were pain (most common), liver dysfunction, anorexia, post-embolization syndrome, ventricular arrhythmia, opiate toxicity, respiratory failure and death.
Efficacy of treatment. Treatment efficacy is summarized in Tables IV, V and VI. Response was initially evaluated at 3 months from the initial DEBIRI treatment. Depending on the degree of response, observation for ongoing response or repeat treatments was performed. The 3-month overall response rate (ORR) (complete response (CR) and partial response (PR)) was 65% for the X-DEBIRI group and 40% for the DEBIRI group, which was statistically beneficial favoring the X-DEBIRI group (p=0.048). However, there were more patients in the DEBIRI group without mRECIST data during the 3-month time interval. The disease control rate (DCR), which is the sum of CR, PR and stable disease (SD), was not statistically significant in the 3-month time interval (Table IV). The efficacy of treatment was not significant between the two groups in 6 months (Table V). During the 12-month evaluation, we found no statistical difference between both groups with regards to ORR. However, there was a statistical difference between both groups in DCR favoring the X-DEBIRI group (X-DEBIRI 73%, DEBIRI 39%; p=0.021).
Survival. Survival curves are shown in Figures 1a, 1b and 1c. OS was measured from the date of first DEBIRI treatment to the date of death and calculated by intent-to-treat (Figure 1a). Based on the Kaplan-Meier curves, median survival was 13 months in the DEBIRI group and 22 months in the X-DEBIRI group, which was not found to be significantly different (log-rank test, p=0.217). Median PFSs were 7 and 9 months for the DEBIRI and X-DEBIRI groups respectively (Figure 1b). Median HPFS were 8 and 12 months for the DEBIRI and X-DEBIRI groups, respectively (Figure 1c); also not statistically significant (log rank; p=0.5). Mean follow-up in all patients was 10 months.
Discussion
Systemic chemotherapy with or without surgery is the treatment-of-choice for patients with metastatic CRC. However, in patients who develop progression or intolerance to therapy, disease control becomes a greater concern. DEBIRI has shown promise in the management of pre-treated or chemotherapy-refractory patients with one prospective trial showing longer median OS and PFS (22 months and 7 months, respectively) in patients treated with DEBIRI compared to the FOLFIRI arm (15 months and 4 months, respectively) (10). Another single-arm, non-randomized study published at our Institution showed similar results in OS and PFS (19 and 11 months, respectively) (18). The response rates in those studies were higher than historic response rates reported for systemic irinotecan in pre-treated patients (10).
On the other hand, the literature on the efficacy of capecitabine in pre-treated patient population has been mixed (11, 13, 14, 23). Capecitabine is an oral fluoropyrimidine carbamate that is converted to 5-FU preferentially in tumors though a three-enzyme cascade. Most pre-treated patients were already FU-resistant; perhaps this is the reason for the overall unfavorable response rate published in the literature.
This study shows that the addition of capecitabine to DEBIRI does not result in additional toxicity than what is expected with DEBIRI alone, which is counter-intuitive given the toxicity profile of capecitabine. Moreover, both groups were similar with regards to the number of high-grade adverse events, which was overall low in both groups (X-DEBIRI 2%; DEBIRI 5%).
In the 3-and 6-month evaluation, there was no difference in the DCR. Curiously, in the 12-month evaluation, there was a statistically favorable DCR in the X-DEBIRI group compared to the DEBIRI group (73% and 39%, respectively). Given the aggressive nature of metastatic treatment refractory CRC, this is felt to represent a more durable anti-tumoral effect with concomitant capecitabine. We believe this finding is quite significant given the fact that our patients had experienced disease progression following prior first-line therapy, most of which were FU-based regimens.
Despite the fact that the median survival, PFS and HPFS were higher for the X-DEBIRI group compared to the DEBIRI group, we did not compute any statistical difference. Since all end-points numerically favored the X-DEBIRI group, and given the suggestion of more durable disease control with concomitant capecitabine, there could be an overall trend towards improved survival. Presence of extrahepatic metastatic disease was an independent prognostic factor and it comprised approximately 30% of each treatment group. We can extrapolate from this data that concomitant capecitabine may address extrahepatic metastatic disease leading to more durable DCR without contributing to increased toxicity. This may have resulted in the demonstrated survival trend.
Given the design of this study, there is inevitable time-dependent assessment bias. We tried to limit this by only including patients within the specified time interval (3 months, 6 months, etc.). This approach limited the bias but decreased our sample size within that timepoint. We favored reducing the bias at the expense of increasing the variability of our data with a smaller sample size. Another limitation is that with this type of study, the two groups cannot be homogenous. There was a relatively small number of patients in the X-DEBIRI group and the samples are uneven limiting the power of the study, especially with regards to survival analysis. Also, the treated radioembolization and DEBDOX proportion of patients were slightly different in each group, which may influence toxicity and efficacy (although they were not found to be independent prognostic factors for survival). Finally, there may have been factors that influenced the administration of concomitant capecitabine (e.g. known 5-FU resistance). Those factors may possibly confound our results.
In conclusion, the addition of concomitant capecitabine to DEBIRI does not add additional toxicity in this pooled cohort. It may also provide a more durable disease control rate. We also found that there may be a trend towards increased survival benefit for the X-DEBIRI group in this small dataset. Further larger studies are warranted to support these findings.
Footnotes
-
Conflicts of Interest
RCGM is a consultant for Biocompatibles. The other authors declare that they have no conflict of interest.
- Received August 16, 2014.
- Revision received September 16, 2014.
- Accepted September 23, 2014.
- Copyright© 2014 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved