Abstract
Background
Lung chemoembolization is an emerging treatment option for lung tumors, but the optimal embolic, drug, and technique are unknown.
Purpose
To determine the technical success rate and safety of bronchial or pulmonary artery chemoembolization of lung metastases using ethiodized oil, mitomycin, and microspheres.
Materials and Methods
Patients with unresectable and unablatable lung, endobronchial, or mediastinal metastases, who failed systemic chemotherapy, were enrolled in this prospective, single-center, single-arm, phase I clinical trial (December 2019–September 2020). Pulmonary and bronchial angiography was performed to determine the blood supply to the lung metastases. Based on the angiographic findings, bronchial or pulmonary artery chemoembolization was performed using an ethiodized oil and mitomycin emulsion, followed by microspheres. The primary objectives were technical success rate and safety, according to the National Cancer Institute Common Terminology Criteria for Adverse Events. CIs of proportions were estimated with the equal-tailed Jeffreys prior interval, and correlations were evaluated with the Spearman test.
Results
Ten participants (median age, 60 years; interquartile range, 52–70 years; six women) were evaluated. Nine of the 10 participants (90%) had lung metastases supplied by the bronchial artery, and one of the 10 participants (10%) had lung metastases supplied by the pulmonary artery. The technical success rate of intratumoral drug delivery was 10 of 10 (100%) (95% CI: 78, 100). There were no severe adverse events (95% CI: 0, 22). The response rate of treated tumors was one of 10 (10%) according to the Response Evaluation Criteria in Solid Tumors and four of 10 (40%) according to the PET Response Criteria in Solid Tumors. Ethiodized oil retention at 4–6 weeks was correlated with reduced tumor size (ρ = –0.83, P = .003) and metabolic activity (ρ = –0.71, P = .03). Pharmacokinetics showed that 45% of the mitomycin dose underwent burst release in 2 minutes, and 55% of the dose was retained intratumorally with a half-life of more than 5 hours. The initial tumor-to-plasma ratio of mitomycin concentration was 380.
Conclusion
Lung chemoembolization was technically successful for the treatment of lung, mediastinal, and endobronchial metastases, with no severe adverse events.
Clinical trial registration no. NCT04200417
© RSNA, 2021
Online supplemental material is available for this article.
See also the editorial by Georgiades et al in this issue.
Summary
Lung chemoembolization can potentially treat large and multifocal lung metastases, including mediastinal lymph nodes, in patients with limited treatment options (ie, chemorefractory, unresectable, and unablatable).
Key Results
■ In a prospective study of 10 participants, all had hypervascular lung metastases that were supplied by the bronchial artery in nine participants (90%) and by the pulmonary artery in one participant (10%).
■ After chemoembolization, the intratumoral mitomycin concentration was 380 times the plasma mitomycin concentration.
■ Drug release from the ethiodized oil and mitomycin emulsion occurred with a half-life of more than 5 hours in vivo and 7.1 hours in vitro.
Introduction
Most patients with lung metastases have unresectable disease (1). Patients with unresectable and unablatable lung metastases have poor survival and limited treatment options when the metastases stop responding to systemic chemotherapy (2).
Transarterial chemoembolization (TACE) of the pulmonary or bronchial arteries is an emerging treatment option for large and multifocal lung tumors (Table E1 [online]). Like the liver, the lung has a dual blood supply. However, unlike liver tumors, which are mostly supplied by the hepatic artery, lung tumors can be supplied by either the bronchial artery or pulmonary artery (3). Response rates are 17% after pulmonary artery chemoembolization (4,5) and 39% after bronchial artery chemoembolization (6–8), bland embolization (9), or chemoinfusion (10–12). The highest reported response rates have been seen with primary lung cancer, treated via the bronchial artery (Table E1 [online]).
A major limitation of existing studies of lung chemoembolization is the low response rate. We hypothesize that the low response rate is because of the dual blood supply of lung tumors. Lung tumors primarily supplied by the pulmonary artery will presumably have a low response rate to bronchial artery chemoembolization and vice versa. Previous studies have shown that lung tumors with more enhancement at arteriography are more likely to respond to chemoembolization, and a thorough search for bronchial and nonbronchial systemic arteries supplying the tumor is critical for achieving a good response after bronchial artery TACE (7,9,10). However, none of the previous studies have examined both the bronchial and pulmonary arteries. Finding the blood supply to the lung metastases will be critical for ensuring adequate tumor coverage.
Using lessons learned from previous studies, we designed a phase I clinical trial of lung chemoembolization for unresectable and unablatable lung metastases that are not responding to chemotherapy. We examined the blood supply of lung metastases by performing both bronchial and pulmonary angiography in all study participants. We performed bronchial or pulmonary artery chemoembolization based on the angiographic findings. Chemoembolization was performed using an ethiodized oil and mitomycin emulsion, followed by particles. Intra-arterial ethiodized oil selectively deposits in lung tumors (4). Mitomycin is preferentially activated under hypoxic conditions (13) and thus should have a synergistic effect with arterial embolization. Intratumoral drug delivery and release were evaluated by measuring ethiodized oil retention in the tumor and plasma concentration of mitomycin at multiple time points after TACE.
This study tested whether lung TACE can be used to treat large and multifocal lung metastases, including hilar lymph nodes and centrally located metastases, in patients with no other good treatment options. The primary objectives were to evaluate the technical success rate and safety of bronchial or pulmonary artery chemoembolization of lung metastases, using ethiodized oil, mitomycin, and microspheres. The secondary objectives included objective response rate, blood supply to the lung metastases (bronchial or pulmonary artery), pharmacokinetics, and ethiodized oil retention.
Materials and Methods
This was a single-center, prospective, phase I, single-arm trial of lung chemoembolization (clinicaltrials.gov identifier: NCT04200417). The trial was approved by the institutional review board at Memorial Sloan-Kettering Cancer Center. All patients provided written informed consent.
Chemoembolization was performed using an ethiodized oil and mitomycin emulsion, followed by spherical particles. No other cancer treatments were given between the preprocedure imaging and the postprocedure imaging to evaluate response. Stopping rules are given in Appendix E1 (online).
This study was funded by the Brockman Medical Research Foundation, the Society of Interventional Oncology, and GE Healthcare. Core facilities were funded in part through a National Institutes of Health and National Cancer Institute Cancer Center Support Grant (no. P30 CA008748). F.E.B. (with 11 years of experience) had control of the data and information submitted for publication. F.E.B. also performed the chemoembolization procedures, performed clinical evaluations, reviewed imaging, and performed the pharmacokinetics analysis. R.Y. (with 11 years of exprience) evaluated response using the Response Evaluation Criteria in Solid Tumors and the PET Response Criteria in Solid Tumors. V.R.T. (with 11 years of experience) performed liquid chromatography–mass spectrometry experiments. M.H. and C.S.M. performed statistical analysis. Data generated or analyzed during the study are available from the corresponding author by request.
Participant Characteristics
Patients were enrolled from December 2019 until September 2020 and were treated from January 2020 until October 2020. This study included patients with unresectable and unablatable lung, endobronchial, mediastinal, or pleural metastases from any primary tumor that were not responding to chemotherapy. The main exclusion criteria were having primary lung cancer, an Eastern Cooperative Oncology Group performance status of 2 or more, oxygen saturation less than 92% on room air, forced expiratory volume in 1 second less than 60%, a pulmonary embolism within 3 months, pneumonia within 1 month, and no measurable lung metastases (eg, only subcentimeter lung nodules).
Preprocedure Evaluation
Within 4 weeks before the chemoembolization procedure, participants were evaluated in clinic and underwent triphasic chest CT (Appendix E1 [online]), fluorine 18 (18F) fluorodeoxyglucose (FDG) PET/CT, spirometry, and laboratory tests (ie, complete blood count, comprehensive metabolic panel, and international normalized ratio).
Ethiodized Oil and Mitomycin Emulsion Preparation
Lyophilized mitomycin C powder (5 mg/m2) (Mutamycin, Accord Healthcare) was dissolved in ethiodized oil (Lipiodol, Guerbet) at 2 mg/mL (maximum ethiodized oil volume, 10 mL). F.E.B. formed a water-in-oil emulsion (14) by adding contrast material (Omnipaque 300, GE Healthcare) (50% of the ethiodized oil and mitomycin volume) using ethiodized-oil-compatible syringes and three-way stopcock. At the beginning of the procedure, the ethiodized oil and mitomycin were pumped through the three-way stopcock at least 20 times to form the emulsion. Immediately before the injection, an additional 20 pumps through the three-way stopcock was performed.
Bronchial and Pulmonary Angiography
Pulmonary and bronchial angiography was performed in all participants to determine the blood supply to the lung metastases. All participants were administered preprocedure antibiotics, typically cefazolin. All procedures were performed under general anesthesia to enable better breath-hold angiography. The femoral artery and vein were both accessed, and bronchial and pulmonary arteriograms were obtained to identify the blood supply to the lung metastases. Pulmonary artery pressure was measured before and after pulmonary artery chemoembolization.
After identifying the blood supply to the target tumors, a helical CT catheter angiogram was obtained in an integrated angiography and CT suite. The scanning delay was 5 seconds after the start of contrast material injection for the bronchial artery and 3 seconds for the pulmonary artery. Chemoembolization was only performed if there was enhancement of the tumor and no enhancement in the esophagus or spinal cord. Enhancement of bronchi or mediastinal lymph nodes was not considered a contraindication to chemoembolization.
Lung Chemoembolization
Chemoembolization was performed via the artery (bronchial, nonbronchial systemic, or pulmonary) that showed the greatest tumor enhancement at angiography. If multiple arteries were supplying the tumor, the mitomycin emulsion was split between the arteries according to the estimated volume of the territory supplied by each artery. However, the pulmonary artery and the bronchial artery (or nonbronchial systemic artery) were not both treated in the same lobe to reduce the risk of lung infarct. For localized disease, chemoembolization was performed as selectively as possible, while still treating the target tumors completely. For multifocal disease, only one lung was treated in a single session for bronchial artery TACE, and only one lobe was treated in a single session for pulmonary artery TACE. For this study, only a single treatment session was allowed.
Chemoembolization was performed through a microcatheter, using an ethiodized oil and mitomycin emulsion, followed by microspheres (Embosphere, Merit Medical). Chemoembolization was performed until there was reduced forward flow and no residual tumor blush (15). If the embolization end point was not reached after delivering the entire mitomycin emulsion, then microspheres were delivered (100–300 µm for the pulmonary artery, and 300–500 µm for the bronchial artery). In five of 10 participants, the bronchial artery was selected using a Mikaelsson catheter (Merit Medical) or Simmons 1 catheter (Merit Medical), a 2-F microcatheter (Progreat, Terumo), and a 0.014-inch microwire (Fathom, Boston Scientific). The ethiodized oil and mitomycin emulsion was injected in 0.2-mL aliquots.
Unenhanced helical CT of the chest was performed in the procedure room at the end of the procedure. Participants were monitored in the postanesthesia care unit for at least 4 hours and then were admitted for overnight observation. A postprocedure neurologic examination was performed.
Postprocedure Evaluation
Laboratory test results (ie, complete blood count and complete metabolic panel) and chest radiographs were obtained the day after the procedure. Participants were evaluated in clinic 1–2 and 4–6 weeks after the procedure. 18F-FDG PET/CT, chest CT, spirometry, and laboratory tests (ie, complete blood count and complete metabolic panel) were performed 4–6 weeks after the procedure.
Technical Success Rate and Safety
Technical success was defined as the delivery of a nonzero amount of ethiodized oil and mitomycin into an artery supplying a target tumor. Complications were classified using the National Cancer Institute Common Terminology Criteria for Adverse Events (version 5.0). The safety assessment period ended 6 weeks after the procedure. Complications were evaluated before discharge after the procedure, at the postprocedure clinic visits, and according to an additional chart review after the second clinic visit.
Treatment Efficacy
Response to treatment was evaluated at PET/CT and chest CT, 4–6 weeks after procedure, using the PET Response Criteria in Solid Tumors (version 1.0) (16) and the Response Evaluation Criteria in Solid Tumors (version 1.1) (17). The response of PET Response Criteria in Solid Tumors is based on the peak standardized uptake value, corrected for lean body mass. The largest treated and largest untreated tumors were measured in each participant.
Ethiodized Oil Retention
Ethiodized oil retention in the largest embolized tumor in each participant was calculated according to Hounsfield units at unenhanced CT performed immediately after the procedure and 4–6 weeks after the procedure (Appendix E1 [online]).
Pharmacokinetics
The plasma mitomycin concentration was measured 0, 10, 20, 40, 80, and 160 minutes after completing chemoembolization. At each time point, 3 mL of venous blood was drawn into a lavender top tube containing ethylenediaminetetraacetic acid and placed on ice. The plasma mitomycin concentration was determined with liquid chromatography–mass spectrometry (Appendix E1 [online]). The plasma mitomycin concentration versus time curve was fit to a pharmacokinetics model (Appendix E1 [online]). Emulsion stability and in vitro drug release experiments were performed (Appendix E1 [online]).
Statistical Analysis
Medians were compared using the Wilcoxon signed-rank test (paired measurements) or Wilcoxon rank-sum test (unpaired measurements). Correlations between continuous variables were evaluated using the Spearman rank correlation coefficient (ρ) and Spearman test. CIs of proportions were estimated using the equal-tailed Jeffreys prior interval (18). P < .05 was considered a statistically significant difference. Statistical analysis was performed with R software (version 3.6.1, R Foundation for Statistical Computing).
Results
Participant Characteristics
Of 23 patients assessed for eligibility, 10 participants (median age, 60 years; interquartile range [IQR], 52–70 years; six women) were allocated to intervention. Most participants in the trial (Fig 1, Table 1) had colorectal cancer (nine of 10 participants [90%]) and were classified as Eastern Cooperative Oncology Group 1 (six of 10 participants [60%]), had at least four treated tumors (six of 10 participants [60%]), and had untreated intrathoracic disease (nine of 10 participants [90%]) and extrathoracic disease (nine of 10 participants [90%]). The median size of the largest measurable treated tumor in each patient was 3.9 cm.
Figure 1:
Flowchart of study enrollment and follow-up.
Table 1:
Participant Characteristics and Treatment
Blood Supply to Lung Metastases
Nine of the 10 participants (90%) had lung metastases supplied by the bronchial artery (ie, all colorectal primary). One participant (10%) had a lung metastasis supplied by the pulmonary artery (metastatic melanoma). Typical examples of bronchial and pulmonary angiograms are shown in Figure 2.
Figure 2:
Diagram of blood supply to lung metastases (arrows). Tumors were classified into three categories, according to their appearance at angiography. TACE = transarterial chemoembolization.
Feeding arteries identified at CT did not predict angiographic findings—four of five participants (80%) with a feeding pulmonary artery identified on a CT angiogram had bronchial artery supply at catheter angiography (Table 2). Mediastinal lymph nodes, endobronchial tumors, and central lung metastases were all supplied by the bronchial artery.
Table 2:
Characteristics of Lung Metastases Supplied by Pulmonary Artery versus Bronchial Artery
In two participants, hypervascular lesions seen at angiography were actually due to focal enhancement of normal pleura at CT catheter angiography (Fig E1 [online]). This highlights the importance of CT catheter angiography to help confirm enhancement of target tumors. In one participant, the pleura was supplied by the bronchial artery, and in the other participant, the pleura was supplied by the pulmonary artery.
Lung Chemoembolization
Bronchial artery chemoembolization was performed in nine participants (90%, Movie [online]), and pulmonary artery chemoembolization was performed in one participant (10%). The technical success rate of intratumoral drug delivery (Fig 3) was 100% (95% CI: 78, 100). In all participants, intratumoral ethiodized oil retention was seen at unenhanced CT immediately after TACE.
Figure 3:
Right bronchial artery chemoembolization in 50-year-old man with colon cancer and growing chemorefractory lung metastases. (A) Right bronchial angiogram shows lung metastases (arrow). (B) Selective right bronchial angiogram shows lung metastases (arrow). (C) Catheter CT angiogram shows blood supply to lung, hilar lymph node, and subcarinal lymph node (arrows). (D) Unenhanced CT scan after transarterial chemoembolization shows retained lipiodol (arrow) in right lung metastases. (E) Pretreatment PET/CT scan shows hypermetabolic lung metastases (arrow). (F) PET/CT scan after transarterial chemoembolization shows partial metabolic response (arrow), with tumor necrosis resulting in intratumoral gas.
Movie:
Fluoroscopy during TACE of a colon cancer lung metastasis (arrows), supplied by the bronchial artery, shows preferential flow and retention of lipiodol in the lung metastasis.
The mean amount of ethiodized oil and mitomycin emulsion delivered in each treatment was 2.3 mL ± 2.0 (standard deviation), of 6.8 mL ± 0.7 that was prescribed, because of reaching the embolization end point (ie, slow flow and no tumor blush). One participant (10%) was administered the full prescribed dose. One participant had an intratumoral arteriovenous fistula (pulmonary artery to vein) that was occluded using microspheres (Embospheres, 300–500 µm) before injecting the ethiodized oil and mitomycin emulsion. Two participants underwent embolization using the microspheres after delivery of the ethiodized oil and mitomycin emulsion. In the one participant who underwent pulmonary artery TACE, the mean pulmonary artery pressure was 13 mm Hg before TACE and 12 mm Hg after TACE.
Safety Analysis
No serious or severe adverse events occurred (95% CI: 0, 22 rate of serious or severe adverse events). All participants met criteria for discharge within 4 hours after the procedure; however, per protocol, they were all monitored overnight and discharged home the next day. Three of the 10 participants (30%) had a grade 1 or 2 adverse event (Table 3). Immediate and 4–6-week postprocedure CT showed changes in normal lung parenchyma in five of the 10 participants (50%)—three had focal ethiodized oil deposition in normal pleura, one had focal ethiodized oil uptake in normal lung parenchyma, and one had a peripheral lung infarct. All 5 participants underwent bronchial artery TACE.
Table 3:
Adverse Events
Median oxygen saturation on room air was 98.5% (IQR, 98%–99%) before the procedure and 98% (IQR, 97%–98%) at 4–6 weeks after the procedure (P = .2). One participant did not undergo postprocedure spirometry because of the limited availability of spirometry during the COVID-19 pandemic. Among the nine participants who underwent both pre- and postprocedure spirometry, median forced vital capacity was 84% (IQR, 70%–96%) before the procedure and 77% (IQR, 73%–88%) at 4–6 weeks after the procedure (P = .02). Median forced expiratory volume in 1 second was 86% (IQR, 73%–93%) before the procedure and 81% (IQR, 70%–88%) at 4–6 weeks after the procedure (P = .05).
Treatment Efficacy
Response of treated metastases at 4–6 weeks according to the Response Evaluation Criteria in Solid Tumors was partial response (one of 10 participants [10%]) and stable disease (nine of 10 participants [90%]). Response of treated tumors at 4–6 weeks according to the PET Response Criteria in Solid Tumors was partial metabolic response (four of 10 participants [40%]), stable metabolic disease (five of 10 participants [50%]), and progressive metabolic disease (one of 10 participants [10%]).
Treated tumors decreased in size, compared with untreated tumors (P = .02). For the nine participants with both treated and untreated tumors, treated tumors were mostly stable to decreased in size after TACE (median change, 0%; IQR: 11%–2%), whereas untreated tumors were mostly increased in size (median change, 10%; IQR: 0%–17%) (Table E3 [online]). We found no evidence of a difference in the percentage change in peak standardized uptake value, corrected for lean body mass for treated versus untreated tumors (P = .1). However, according to the PET Response Criteria in Solid Tumors criteria, four of the 10 treated tumors (40%) demonstrated a metabolic response to treatment, and 0 of the 10 untreated tumors (0%) demonstrated a metabolic response. The untreated tumors served as an internal control—they have the same pathologic characteristics, in the same participant, at the same point in time.
Participants with more intratumoral ethiodized oil retention at 4–6 weeks showed a greater decrease in tumor size (ρ = –0.83, P = .003). We did not find evidence of change in tumor size at 4–6 weeks being correlated with any of the following factors: (a) initial tumor size, (b) distance to hilum or pleura, (c) amount of mitomycin delivered, (d) use of microspheres, (e) ethiodized oil retention in the tumor immediately after TACE, (f) mitomycin tumor-to-plasma ratio, or (g) complications (Tables E4, E6, E7 [online]). Results should be interpreted cautiously because of the small sample size.
Participants with more intratumoral ethiodized oil retention at 4–6 weeks (ρ = –0.71, P = .03), greater distance to pleura (ρ = –0.64, P = .04), or adverse events (P = .03) had a greater decrease in standardized uptake value, corrected for lean body mass. We did not find evidence of change in standardized uptake value, corrected for lean body mass at 4–6 weeks, being correlated with any of the following factors: (a) initial tumor size, (b) distance to hilum, (c) amount of mitomycin delivered, (d) use of microspheres, (e) ethiodized oil retention in the tumor immediately after TACE, or (f) mitomycin tumor-to-plasma ratio (Tables E5–E7 [online]).
Ethiodized Oil Retention
Immediately after TACE, the intratumoral ethiodized oil concentration was 1.0% (vol/vol; geometric mean; range, 0.51%–2.1%). Given the initial mitomycin concentration dissolved in ethiodized oil, this translates to an intratumoral mitomycin concentration of 21000 ng/mL (geometric mean; range, 10000–41000 ng/mL) compared with a plasma mitomycin concentration of 54 ng/mL (geometric mean; range, 7.3–270 ng/mL). The initial tumor-to-plasma ratio of mitomycin concentration was 380 (geometric mean; range, 69–3300).
The mean amount of residual ethiodized oil in the tumor at 4–6 weeks was 27% (range, 0%–77%) of the initial amount. Assuming first-order kinetics, this corresponds to a half-life of 16 days for intratumoral ethiodized oil retention. The half-life was 28 days in participants where microspheres were used and 13 days in participants where microspheres were not used.
Pharmacokinetics Analysis
A previous publication showed that after administration of intravenous mitomycin, the plasma concentration immediately started to decrease, following a standard two-compartment pharmacokinetics model (19). The median initial plasma concentration in nanograms per meter, normalized by the mitomycin dose in milligrams per meter squared, was 119 (ng/mL)/(mg/m2).
After lung TACE using the ethiodized oil and mitomycin emulsion, there were two main differences in the pharmacokinetics curves (Fig 4), compared with intravenous mitomycin. First, the initial plasma concentration was lower (51 [ng/mL]/[mg/m2]; P = .004). Second, the plasma concentration sometimes increased or plateaued, rather than decrease immediately after finishing chemoembolization.
Figure 4:
Graph shows plasma concentration of mitomycin after chemoembolization of lung metastases, measured with liquid chromatography–mass spectrometry.
Both of these differences can be explained by a model that includes initial burst release, as well as sustained release of chemotherapy retained in the tumor (Figs 5–E2 [online]). The pharmacokinetics curve has three phases. First, burst release of mitomycin after TACE (45% of the dose, half-life of 2 minutes) results in briefly increasing mitomycin plasma concentration as the drug is released. Because only a fraction of the drug undergoes burst release, the plasma concentration is lower than intravenous delivery. Second, systemic clearance of the burst-released drug occurs, with the same half-life as intravenous delivery. Third, the drug that was retained in the tumor is slowly released (55% of the dose, half-life >330 minutes). The third phase was not directly measured in this study because the last time point was at 160 minutes; thus, we can only provide a lower bound on the half-life of sustained drug release from tumor. To address this limitation, in vitro drug release experiments were performed.
Figure 5:
(A) Pharmacokinetics model for transarterial chemoembolization, with parameters (means ± standard deviations) fit to experimental data in Figure 4. Arrows are color coded to match labels in B and C. (B) Graph shows plasma mitomycin concentration (log scale), using parameters from A. (C) Graph shows mitomycin retained in tumor (log scale), using parameters from A.
Emulsion Stability and in Vitro Drug Release
The ethiodized oil and mitomycin emulsion was stable in vitro for 5 days, then it completely separated 3 days later (Fig 6, Appendix E1 [online]). Drug release from the ethiodized oil and mitomycin emulsion occurs before the emulsion separates— 50% of the drug is released in 7.1 hours, whereas 50% of the emulsion separates in 6.2 days.
Figure 6:
Emulsion separation and drug release in vitro. (A) Photographs of lipiodol and mitomycin emulsion (top syringe) and separated emulsion (bottom syringe). In separated emulsion, lipiodol is top layer and aqueous phase is bottom layer. (B) Graph shows emulsion separation kinetics. Volume of aqueous phase was fit to cumulative distribution function of gamma distribution (α = 69, β = 0.10 days). (C) Graph shows drug release kinetics.
Discussion
Lung chemoembolization enables treatment of large and multifocal lung metastases, including mediastinal lymph nodes, in patients with limited treatment options. There were no serious adverse events, and the metabolic response rate was 40% in patients with chemorefractory cancer who were not receiving any other cancer therapy.
The pulmonary versus bronchial blood supply to lung metastases has not previously been well established in the literature. A previous study from 1987, using ex vivo angiography of resected lung metastases, suggested that all lung metastases are supplied by the pulmonary artery, with a minor contribution from the bronchial artery (3). Here, using modern imaging equipment, and in vivo angiography, we found that 90% of study participants had lung metastases supplied by the bronchial artery, and 10% were supplied by the pulmonary artery. Certain types of lung lesions might be more likely to be supplied by the pulmonary artery: pleural lesions and noncolorectal and nonlung histologic findings. A previous study showed that metastases from noncolorectal primaries have a much higher response rate after pulmonary artery chemoembolization, compared with colorectal metastases (4), suggesting that noncolorectal metastases can be supplied by the pulmonary artery. The highest reported response rates have been seen with primary lung cancer, treated via the bronchial artery, suggesting that primary lung cancer is supplied by the bronchial artery.
Unlike the liver, where metastases can be either hypervascular or hypovascular at angiography compared with normal liver, in the lung, all participants in this trial had hypervascular metastases at either bronchial or pulmonary angiography. This makes the intra-arterial route an attractive option for drug delivery into lung metastases.
The ideal drug carrier for TACE has two main features—specific drug delivery into the tumor compared with normal tissue and sustained intratumoral release of a therapeutic concentration of a drug. We evaluated tumor targeting and drug release of the ethiodized oil and mitomycin emulsion using pharmacokinetics, ethiodized oil retention in Hounsfield units, and in vitro drug release experiments. These experiments showed good tumor targeting and drug release and also highlighted some avenues for future improvement.
Intratumoral ethiodized oil retention was seen at CT after TACE in all cases, and minimal ethiodized oil deposition was visible in normal lung parenchyma. The initial tumor-to-plasma mitomycin concentration ratio was 380, and this high intratumoral drug concentration enabled treatment of tumors that were refractory to systemic chemotherapy. The half-life of intratumoral ethiodized oil retention was 16 days. Thus, ethiodized oil serves as a tumor-targeting drug carrier. Similarly, intra-arterial ethiodized oil is also specifically retained in hepatocellular carcinoma (20), colorectal liver metastases (21), other liver tumors (22,23), and also in tumors outside the liver such as pancreatic (24,25) and renal (26) tumors. Ethiodized oil appears to specifically localize in tumors versus normal tissue but is nonspecific for the type of tumor.
The main limitations of ethiodized oil as an intra-arterial drug carrier are the heterogeneous uptake and early washout seen in some tumors. Poor ethiodized oil uptake or retention after lung TACE was associated with lower response rates in our study, similar to what has been seen after hepatic artery TACE (23). Experimental drug carriers with improved tumor uptake and specificity (27) could result in more effective local drug delivery.
The kinetics of intratumoral drug release after lung chemoembolization could be another important variable affecting response rates. Prolonged exposure to chemotherapy can result in greater tumor necrosis (28) because a subpopulation of tumor cells are transiently resistant to chemotherapy at any given moment (29). Thus, we anticipate that sustained intratumoral release of a drug will be important for optimal response. In our study, the pharmacokinetics model showed that the intratumoral drug retention half-life was more than 5 hours, and the in vitro experiments showed a 50% drug release in 7.1 hours. Drug release occurred before emulsion separation, likely because mitomycin is soluble in both water and ethiodized oil. A purely water-soluble drug would not be released until the water-in-oil emulsion separates, which occurred in 6.2 days in vitro (50% separation). Improved ethiodized oil emulsions, which have less burst release, and longer drug release kinetics (30), could improve outcomes.
Limitations of this trial are primarily related to the small number of participants, the lack of a control group, and the fact that only one treatment was allowed per participant. Only one participant had metastases from a noncolorectal primary, and only one participant was treated via the pulmonary artery. Intratumoral drug retention and release were estimated but were not directly measured.
Currently, we recommend that lung chemoembolization should only be performed as part of a clinical trial. Based on the results of this trial, we suggest the following for future trials: (a) although triphasic chest CT was not helpful, CT angiography is helpful for identifying the anatomy of bronchial and nonbronchial systemic arteries; (b) bronchial angiography could be performed first, and pulmonary angiography could be reserved for patients in which the bronchial artery does not supply the lung tumors; (c) particles should be injected after ethiodized oil and mitomycin and before reaching stasis, even if the entire mitomycin dose cannot be delivered; (d) other cancer therapies, including systemic therapy and liver-directed therapy, could be combined with lung TACE; and (e) if staged lung TACE procedures are performed, spirometry should be performed before and after each procedure, given the changes in forced vital capacity and forced expiratory volume in 1 second seen after lung TACE.
In conclusion, our data suggested that lung chemoembolization can safely treat lung, mediastinal, and endobronchial metastases, with minimal systemic toxicity. High intratumoral drug concentrations after transarterial chemoembolization can overcome chemoresistance. Data on blood supply to lung metastases, and tumor targeting and kinetics of local drug release, will help design future trials. Larger trials should be undertaken to confirm our results regarding safety and efficacy.
Acknowledgments
Acknowledgments
Elena N. Petre, Lynn A. Brody, Timothy W.I. Clark, and Ciara Kelly provided helpful comments on the study design. Rocio P. Johnston and Jimmy Chin protocoled and supervised the triphasic CT scans. Raphael Doustaly of GE Healthcare helped with image processing. Michelle Clark and Cheyenne Samson were the clinical research coordinators for the study. Mark G. Klang determined mitomycin solubility and stability. Finally, we would like to thank the patients, family members, and staff who participated in the study.
Supported by the Brockman Medical Research Foundation, Society of Interventional Oncology, and GE Healthcare. Core facilities were funded in part through a National Institutes of Health and National Cancer Institute Cancer Center Support Grant (no. P30 CA008748).
Disclosures of Conflicts of Interest: F.E.B. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: received payment for lectures from the Society of Interventional Oncology; holds stock/stock options in Claripacs, Labdoor, Qventus, CloudMedx, Notable Labs, and Xgenomes; received reimbursement from Guerbet for travel, accommodations, and meeting expenses; received research support from GE Healthcare, Bayer, Steba Biotech, and Terumo. Other relationships: has patents pending and issued. N.E.K. disclosed no relevant relationships. C.T.S. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: is a consultant for Sirtex, Terumo, Varian, Boston Scientific/BTG, Ethicon; institution has grants/grants pending with Sirtex, Ethicon, Boston Scientific/BTG, National Institutes of Health; receives payment for lectures, including service on speakers bureaus, from Ethicon; receives payment from Ethicon for development of educational presentations; receives reimbursement from Ethicon, Terumo, and BTG for travel, accommodations, and meeting expenses. Other relationships: disclosed no relevant relationships. R.Y. disclosed no relevant relationships. V.R.T. disclosed no relevant relationships. M.H. disclosed no relevant relationships. C.S.M. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: has grants/grants pending with the National Institutes of Health and the Foundation for the National Institutes of Health; receives reimbursement from RSNA for travel, accommodations, and meeting expenses. Other relationships: disclosed no relevant relationships. E.Z. Activities related to the present article: institution has grant with Guerbet. Activities not related to the present article: has research grants/grants pending. Other relationships: disclosed no relevant relationships. H.Y. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: has grants/grants pending with the Thompson Foundation. Other relationships: disclosed no relevant relationships. A.B. disclosed no relevant relationships. S.B.S. disclosed no relevant relationships.
Abbreviations:
- IQR
- interquartile range
- TACE
- transarterial chemoembolization
References
- 1. Moorcraft SY , Ladas G , Bowcock A , Chau I . Management of resectable colorectal lung metastases . Clin Exp Metastasis 2016. ; 33 ( 3 ): 285 – 296 . [DOI] [PubMed] [Google Scholar]
- 2. Sartore-Bianchi A , Amatu A , Bonazzina E , et al . Pooled Analysis of Clinical Outcome of Patients with Chemorefractory Metastatic Colorectal Cancer Treated within Phase I/II Clinical Studies Based on Individual Biomarkers of Susceptibility: A Single-Institution Experience . Target Oncol 2017. ; 12 ( 4 ): 525 – 533 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Milne EN , Zerhouni EA . Blood supply of pulmonary metastases . J Thorac Imaging 1987. ; 2 ( 4 ): 15 – 23 . [DOI] [PubMed] [Google Scholar]
- 4. Vogl TJ , Lehnert T , Zangos S , et al . Transpulmonary chemoembolization (TPCE) as a treatment for unresectable lung metastases . Eur Radiol 2008. ; 18 ( 11 ): 2449 – 2455 . [DOI] [PubMed] [Google Scholar]
- 5. Vogl TJ , Mekkawy AIA , Thabet DB , et al . Transvenous pulmonary chemoembolization (TPCE) for palliative or neoadjuvant treatment of lung metastases . Eur Radiol 2019. ; 29 ( 4 ): 1939 – 1949 . [DOI] [PubMed] [Google Scholar]
- 6. Kennoki N , Hori S , Yuki T , Hori A . Transcatheter Arterial Chemoembolization with Spherical Embolic Agent in Patients with Pulmonary or Mediastinal Metastases from Breast Cancer . J Vasc Interv Radiol 2017. ; 28 ( 10 ): 1386 – 1394 . [DOI] [PubMed] [Google Scholar]
- 7. Hori A , Ohira R , Nakamura T , et al . Transarterial chemoembolization for pulmonary or mediastinal metastases from hepatocellular carcinoma . Br J Radiol 2020. ; 93 ( 1110 ): 20190407 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Zeng Y , Yin M , Zhao Y , et al . Combination of Bronchial Arterial Infusion Chemotherapy plus Drug-Eluting Embolic Transarterial Chemoembolization for Treatment of Advanced Lung Cancer-A Retrospective Analysis of 23 Patients . J Vasc Interv Radiol 2020. ; 31 ( 10 ): 1645 – 1653 . [DOI] [PubMed] [Google Scholar]
- 9. Seki A , Hori S , Sueyoshi S , et al . Transcatheter arterial embolization with spherical embolic agent for pulmonary metastases from renal cell carcinoma . Cardiovasc Intervent Radiol 2013. ; 36 ( 6 ): 1527 – 1535 . [DOI] [PubMed] [Google Scholar]
- 10. Nakanishi M , Demura Y , Umeda Y , et al . Multi-arterial infusion chemotherapy for non-small cell lung carcinoma--significance of detecting feeding arteries and tumor staining . Lung Cancer 2008. ; 61 ( 2 ): 227 – 234 . [DOI] [PubMed] [Google Scholar]
- 11. Yuan Z , Li WT , Ye XD , Dong S , Peng WJ . Intra-arterial infusion chemotherapy for advanced non-small-cell lung cancer: preliminary experience on the safety, efficacy, and clinical outcomes . J Vasc Interv Radiol 2013. ; 24 ( 10 ): 1521 – 1528 . [DOI] [PubMed] [Google Scholar]
- 12. Duan F , Wang MQ , Liu FY , Wang ZJ , Song P , Wang Y . Sorafenib in combination with transarterial chemoembolization and bronchial arterial chemoinfusion in the treatment of hepatocellular carcinoma with pulmonary metastasis . Asia Pac J Clin Oncol 2012. ; 8 ( 2 ): 156 – 163 . [DOI] [PubMed] [Google Scholar]
- 13. Kennedy KA , Rockwell S , Sartorelli AC . Preferential activation of mitomycin C to cytotoxic metabolites by hypoxic tumor cells . Cancer Res 1980. ; 40 ( 7 ): 2356 – 2360 . [PubMed] [Google Scholar]
- 14. de Baere T , Arai Y , Lencioni R , et al . Treatment of Liver Tumors with Lipiodol TACE: Technical Recommendations from Experts Opinion . Cardiovasc Intervent Radiol 2016. ; 39 ( 3 ): 334 – 343 . [DOI] [PubMed] [Google Scholar]
- 15. Jin B , Wang D , Lewandowski RJ , et al . Chemoembolization endpoints: effect on survival among patients with hepatocellular carcinoma . AJR Am J Roentgenol 2011. ; 196 ( 4 ): 919 – 928 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. O JH , Lodge MA , Wahl RL . Practical PERCIST: A Simplified Guide to PET Response Criteria in Solid Tumors 1.0 . Radiology 2016. ; 280 ( 2 ): 576 – 584 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Eisenhauer EA , Therasse P , Bogaerts J , et al . New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1) . Eur J Cancer 2009. ; 45 ( 2 ): 228 – 247 . [DOI] [PubMed] [Google Scholar]
- 18. Brown LD , Cai TT , DasGupta A , et al . Interval estimation for a binomial proportion - Comment - Rejoinder . Stat Sci 2001. ; 16 ( 2 ): 101 – 133 . [Google Scholar]
- 19. den Hartigh J , McVie JG , van Oort WJ , Pinedo HM . Pharmacokinetics of mitomycin C in humans . Cancer Res 1983. ; 43 ( 10 ): 5017 – 5021 . [PubMed] [Google Scholar]
- 20. Zheng XH , Guan YS , Zhou XP , et al . Detection of hypervascular hepatocellular carcinoma: Comparison of multi-detector CT with digital subtraction angiography and Lipiodol CT . World J Gastroenterol 2005. ; 11 ( 2 ): 200 – 203 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Gruber-Rouh T , Naguib NN , Eichler K , et al . Transarterial chemoembolization of unresectable systemic chemotherapy-refractory liver metastases from colorectal cancer: long-term results over a 10-year period . Int J Cancer 2014. ; 134 ( 5 ): 1225 – 1231 . [DOI] [PubMed] [Google Scholar]
- 22. de Baere T , Zhang X , Aubert B , et al . Quantification of tumor uptake of iodized oils and emulsions of iodized oils: experimental study . Radiology 1996. ; 201 ( 3 ): 731 – 735 . [DOI] [PubMed] [Google Scholar]
- 23. Miszczuk MA , Chapiro J , Geschwind JH , et al . Lipiodol as an Imaging Biomarker of Tumor Response After Conventional Transarterial Chemoembolization: Prospective Clinical Validation in Patients with Primary and Secondary Liver Cancer . Transl Oncol 2020. ; 13 ( 3 ): 100742 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Yarmohammadi H , Gonzalez-Aguirre AJ , Schook L , et al . Treatment of pancreatic cancer by intra-arterial injection of an emulsion of lipiodol and bumetanide (an anti-glycolytic drug) in a porcine model: Initial results . J Vasc Interv Radiol 2017. ; 28 ( 2 ): S8 – S9 . [Google Scholar]
- 25. Yoshida H , Onda M , Tajiri T , et al . Experience with intraarterial infusion of styrene maleic acid neocarzinostatin (SMANCS)-lipiodol in pancreatic cancer . Hepatogastroenterology 1999. ; 46 ( 28 ): 2612 – 2615 . [PubMed] [Google Scholar]
- 26. Tsuchiya K , Uchida T , Kobayashi M , Maeda H , Konno T , Yamanaka H . Tumor-targeted chemotherapy with SMANCS in lipiodol for renal cell carcinoma: longer survival with larger size tumors . Urology 2000. ; 55 ( 4 ): 495 – 500 . [DOI] [PubMed] [Google Scholar]
- 27. Nurili F , Yarmohammadi H , Bendet A , et al . Improved intra-arterial tumor targeting using an omniphobic oil in a pig model . J Vasc Interv Radiol 2019. ; 30 ( 3S ): S204 . [Google Scholar]
- 28. Baguley BC , Marshall ES , Whittaker JR , et al . Resistance mechanisms determining the in vitro sensitivity to paclitaxel of tumour cells cultured from patients with ovarian cancer . Eur J Cancer 1995. ; 31A ( 2 ): 230 – 237 . [DOI] [PubMed] [Google Scholar]
- 29. Spencer SL , Gaudet S , Albeck JG , Burke JM , Sorger PK . Non-genetic origins of cell-to-cell variability in TRAIL-induced apoptosis . Nature 2009. ; 459 ( 7245 ): 428 – 432 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Deschamps F , Isoardo T , Denis S , et al . Biodegradable Pickering emulsions of Lipiodol for liver trans-arterial chemo-embolization . Acta Biomater 2019. ; 87 ( 177 ): 186 . [DOI] [PubMed] [Google Scholar]
- 31. Boas FE , Brody LA , Erinjeri JP , et al . Quantitative Measurements of Enhancement on Preprocedure Triphasic CT Can Predict Response of Colorectal Liver Metastases to Radioembolization . AJR Am J Roentgenol 2016. ; 207 ( 3 ): 671 – 675 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Borgheresi A , Gonzalez-Aguirre A , Brown KT , et al . Does Enhancement or Perfusion on Preprocedure CT Predict Outcomes After Embolization of Hepatocellular Carcinoma? Acad Radiol 2018. ; 25 ( 12 ): 1588 – 1594 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Boas FE , Kamaya A , Do B , et al . Classification of hypervascular liver lesions based on hepatic artery and portal vein blood supply coefficients calculated from triphasic CT scans . J Digit Imaging 2015. ; 28 ( 2 ): 213 – 223 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Lide DR . CRC Handbook of Chemistry and Physics . 81st ed. New York, NY: : CRC Press, 2000. . [Google Scholar]
- 35. Belikov AV . The number of key carcinogenic events can be predicted from cancer incidence . Sci Rep 2017. ; 7 ( 1 ): 12170 . [DOI] [PMC free article] [PubMed] [Google Scholar]