Immuno-Fluorescent Assessment of Ablated Colorectal Liver Metastases: The Frozen Section of Image-Guided Tumor Ablation?

Abstract

Purpose

To validate an immuno-fluorescent assay (IFA) detecting residual viable tumor as intraprocedural thermal ablation (TA) zone assessment and demonstrate its prognostic value for local tumor progression after colorectal liver metastases (CLM) TA.

Materials and Methods

This prospective, IRB-approved study included 99 patients with 155 CLMs ablated between November 2009 and January 2019. Tissue samples from the ablation zone (AZ) center and minimal margin underwent immunofluorescent microscopic examination interrogating cellular morphology and mitochondrial viability (IFA) within 30 minutes after ablation. The same tissue samples were subsequently evaluated with standard morphological and immunohistochemical (IHC) methods. Sensitivity, specificity, and overall accuracy of IFA versus standard morphological and IHC examination were calculated. Local tumor progression (LTP)-free survival rates were evaluated for 12-month follow-up period.

Results

Of the 311 tissue samples stained, 304 (98%) were deemed evaluable. 27% (81/304) of specimens were considered positive for the presence of viable tumor. The accuracy of IFA was 94% (286/304). Sensitivity and specificity were 100% (63/63) and 93% (223/241), respectively. The 18 false-positive IFA assessments corresponded to samples that included viable cholangiocytes. The 12-month LTP-free survival was 59% vs 78% for IFA positive vs negative for viable tumor AZs, respectively (P <.001). There was no difference in LTP between margin positive only versus central AZ positive tumors (25% vs 31%, p=1).

Conclusion

IFA assessment of the AZ can be completed intra-procedurally and serve as a valid real-time biomarker of complete tumor eradication or detect residual viable tumor after TA. This method could improve tumor control by TA.

Introduction

The main limitation for the widespread use of percutaneous image-guided thermal ablation (TA) has been incomplete treatment and local tumor progression (LTP) [1–4]. The spatial resolution of post-ablation imaging cannot detect residual small volume, viable tumor (VT) cells [5]. Consequently, LTP occurs even in the face of complete tumor ablation with sufficient margins by imaging [2, 3, 6–9]. Real-time prognostic markers identifying patients at increased risk for LTP that could guide decisions regarding additional ablations within the same session are necessary [10, 11].

Unlike surgical resection, TA is not accompanied by pathological assessment of the treated tumor and surrounding margins. VT in tissue adherent to radiofrequency electrodes after ablation was associated with shorter LTP-free, cancer-specific, and overall survival [12–15]. Biopsy of the center and margin of the ablation zone (AZ) with morphological and immunohistochemical assessments highly predicted LTP and LTP-free survival in a prospective study [6]. Hence, post-ablation tissue examinations can be an independent prognosticator of oncological outcomes after TA [12–15].

Routine processing of AZ biopsy specimens required days, precluding immediate evaluation at the time of the procedure[12–15]. The immuno-fluorescent assessment (IFA) of ablated liver tissue has been proposed as a real-time evaluation method of tissue morphology and viability since it can be completed within 30 minutes after sampling [16]. Unlike routine on-site microscopy that only evaluates cellular morphology, IFA has the ability to interrogate mitochondrial viability and in combination with the fluorescent morphologic assessment can distinguish viable from non-viable tumor cells. The purpose of this single-arm, prospective study was to evaluate the sensitivity and specificity of IFA in assessing TA completeness and describe its correlation with LTP [6, 16].

Materials and Methods

Patient Selection

This IRB-approved prospective, NIH-funded single-arm study was designed to assess ablated colorectal cancer liver metastases (CLM) through biopsy of the AZ center and margin, followed by IFA. All patients with CLM undergoing TA with local curative intent were eligible for inclusion in the study. The study did not alter the inclusion/exclusion criteria for ablation of CLM, offered as standard of care, according to institutional and IR service guidelines. These clinical guidelines changed over time; initially tumors up to 5 cm were referred to and accepted for ablation but adopting to general trends the largest diameter tumor acceptable was decreased to 3 cm.

Treatment

TA was administered per standard of care to patients with CLM within this study. The added study intervention was biopsy of the AZ center and margin in order to detect residual VT, after imaging depiction of complete tumor ablation with the desired margin (minimum of 5 mm). All procedures were performed under general anesthesia with radiofrequency (RFA) or microwave (MWA) ablation depending on tumor characteristics and operator preference. Applicator placement and tumor targeting were conducted under CT guidance. CT-fluoroscopy (100%), real-time PET/CT (72%), and ultrasound guidance (35%) were used as needed. Overlapping ablations were performed to create an AZ with a minimum ablation margin of ≥ 5mm around the target tumor [17]. Of note, the IR physicians performing the ablations were blinded to the results of the IFA; no modifications were made to the treatment plan based on the biopsy result.

Margin Assessment

Following ablation, an immediate contrast enhanced CT (CECT) assessed the AZ, with ablation margins measured using a previously described 2D method [7]. The minimal margin (MM) was assessed and measured by comparing the portal venous phases of the immediate post-ablation and pre-ablation CECT (and/or PET/CT). Fusion methods allowing superimposition of the pre-ablation tumor (on PET/CT or CECT) on the post-ablation CECT were also utilized to better depict the MM (Figure 1).

Figure 1.

42-year old male with metastatic colon adenocarcinoma to the liver, status post resection of the primary tumor, intrahepatic and systemic chemotherapy, multiple liver ablations and hepatic resections, presenting with a new 1.7 × 1.7 cm metastasis abutting the inferior vena cava. Preablation CECT (A) and PET/CT (B). Immediate post ablation CECT portal venous phase CECT shows the ablation zone (C). Fusion of immediate post-ablation portal venous phase CECT with the pre-ablation PET/CT (D), used to depict the ablation zone relative to where the tumor resided (white arrow) and the minimal ablation margin (black arrow). This allows accurate sampling of the ablation zone center (E) and minimal margin (F).

Biopsy and Tissue Analysis

Immediately after ablation, 18–20 gauge core biopsy specimens were collected from the AZ center and the MM [17]. At least one core was obtained from the center and one core from the minimal margin of the AZ. For smaller tumors and when the sample could include the center and the MM, one sample was allowed. When present, tissue adherent to the applicators was also collected [12, 18]. All tissue samples were submitted for IFA and subsequently fixed for further morphological analysis with hematoxylin and eosin (H&E) and immunohistochemistry (IHC) for viability (OX-PHOS AB) and proliferation (Ki67) markers. Please refer to the supplemental material for further details regarding the technique for specimen processing.

The study pathologist evaluated the IFA images for the presence of VT cells. Cancer cells were identified by densely distributed nuclei in Hoechst fluorophore. When these cells were also positive for MT Red (mitochondrial viability marker), the sample was classified as VT (Figure 2). The entire specimen and corresponding AZ were categorized as VT even if only one tumor cell was considered viable. Hoechst staining of characteristically large round nuclei (normal hepatocytes) with matching strong MT Red staining resulted in classification of the sample as normal liver tissue (Figure 3). Lack of MT Red signal in regions containing cells identified by nuclear Hoechst, resulted in sample classification as necrotic tissue-thermal artifact (Figure 4).

Figure 2.

(2a) Hoechst stain, a nuclear fluorescent marker equivalent to H&E stain, depicts viable tumor cells in blue. (2b) MT stain, mitochondrial viability fluorescent marker, labels active mitochondria in live cells in red. (2c) Composite image, in which the Hoechst stain is added to MT for counterstaining and labeling of all nuclei in the sample. Cells that are positive for both Hoechst and MT stains are classified as viable tumor cells.

Figure 3.

(3a) Hoechst stain depicts normal hepatocytes in blue. (3b) MT stain labels active mitochondria in live cells in red. (3c) Composite image in which the Hoechst stain is added to MT for counterstaining.

Figure 4.

Necrotic tissue defined by the absence of MT (red) in regions stained with Hoescht (green) stain, composite image.

Following the recording of fluorescent signals, the same tissue samples were fixed in formalin, embedded in paraffin, sectioned, and stained with H&E. The study pathologist, blinded to their own IFA interpretation, classified the same tissue samples as tumor, normal liver tissue, or coagulation necrosis. Tumor specimens were further assessed by IHC for nuclear proliferation with marker Ki67 and mitochondrial viability with marker OxPhos antibody (OXP) [12–16, 19, 20]. This allowed further classification of tumor cells into viable or necrotic. The morphological H&E classification combined with the Ki67 and OXP IHC served as the gold standard reference for evaluation of the AZ and final classification of tissue samples [6, 16].

Imaging Follow-Up

After the immediate post-ablation CECT, TA efficacy was assessed with a triphasic CECT after 4–8 weeks [21]. Subsequent CECT was performed at 2–4-month intervals for up to 12 months after ablation. Additional imaging with PET/CT or MRI was used as needed, for the majority of the patients. All imaging was prospectively assessed by study-dedicated faculty. Evidence of tumor progression within 1 cm from the AZ on CT was considered LTP [21].

Statistical Analysis

Each ablated tumor was considered an independent event. The sensitivity, specificity, and overall accuracy of IFA were calculated using descriptive statistics. LTP-free survival was estimated using the Kaplan-Meier method and log-rank test. The presence of at least one tissue sample positive for VT led to the classification of the entire AZ as positive. Data were analyzed with statistical software (Stata, version 12.0, Tex).

Results

Treatment

Between November 2009 and January 2019, 155 CLM in 99 patients underwent TA followed by biopsy of the AZ and IFA (Table 1). 84% (130/155) of patients had undergone prior liver resection and 94% (145/155) prior chemotherapy. RFA was used in 83/155 (54%) CLM and MWA in 72/155 (46%).

Table 1.

Patient and tumor characteristics

Patient (n=99) and Tumor (n=155) Characteristics
Characteristic	Value
Age (y)*	58 (32–82)
Sex
Female	40 (40)
Male	59 (60)
Tumor size (cm)**	2.0 (0.6–4.6)
LN status at staging of primary disease
Positive	66 (67)
Negative	33 (33)
Synchronous CLM	71 (72)
Time between diagnosis of colorectal cancer and ablation (mo)*	41 (2–151)
No. of tumors treated per patient within protocol**	1.6 (1–8)

Unless otherwise indicated, data represent the number of patients, and data in parentheses are percentages.

^*Data are median values, and data in parentheses represent the range.

^**Data are mean values, and data in parentheses represent the range.

Tissue examination

311 post-ablation tissue samples were collected from the 155 ablated CLMs (mean: 2.0 samples/AZ). Specifically, 138/311 samples were obtained from the AZ center, 138/311 from the suspected minimal margin, and 35/311 from tissue adherent to the ablation applicators.

7/311 samples were unsuccessfully stained with fluorescent stains due to inappropriate incubation time (Figure 5). Using IFA, 81 (27%) of the 304 adequate specimens were considered positive for the presence of VT. Of these, morphologic H&E assessment showed that 63/81 (78%) did in fact contain VT. The eighteen false-positive samples classified as VT by IFA corresponded to preserved cholangiocytes lining small bile ducts. The cholangiocyte distribution around the bile ducts was misinterpreted on the immediate fluorescent images with Hoechst nuclear stain as indicating the glandular morphology of adenocarcinoma (Figure 6). All 223 specimens (100%) classified as necrotic or normal tissue at IFA displayed characteristics of complete coagulative necrosis/thermal artifact and/or normal liver tissue on standard H&E and did not undergo further processing with IHC.

Figure 5.

Schematic diagram of tissue evaluated by immediate fluorescent assessment.

Figure 6.

Example of a false-positive sample identified as tumor cells on the fluorescent images (6a), that in fact corresponded to preserved cholangiocytes lining small bile ducts on H&E (6b).

The overall accuracy of IFA for the detection of tumor vs coagulation necrosis/thermal artifact and/or normal liver tissue was 94% (286/304). Sensitivity for identifying VT was 100% (63/63), and specificity was 93% (223/241). Of the 63 specimens showing tumor on H&E, 50 contained adequate amount of tissue and could be further processed with IHC. Two specimens were positive only for Ki67, and 48 specimens were positive for both Ki67 and OXP markers confirming the presence of VT. There was no significant difference in the incidence of VT detected by IFA in the CLMs ablated with MWA 35/72 [49%] vs RFA 29/83 [35%], (p =0.102). Similarly, there was no association between IFA VT-positive AZs and tumor size > 3cm (n=24), (p=1).

LTP-free survival

64/155 (41%) AZ were classified as positive for VT by IFA. Technical efficacy, as evaluated by the post-ablation CECT at 4–8 weeks, was 98% (152/155); the six specimens originating from the 3 incompletely treated AZs were positive for VT on both IFA and standard evaluations. At 12-month follow-up (median 12 months, range 1–12), LTP occurred in 24/64 (38%) AZs classified by IFA as VT, and in 13/91 (14%) AZs classified as negative (P<.001). The corresponding 12-month LTP-free survival was 59% vs 78%, respectively. (P<.001). Kaplan-Meier survival curves are displayed in Figure 7, Table 2. There were 47 tumors that had only one positive specimen, 35/47 from the center and 12/47 from the margin. Of the 35 tumors with only one positive specimen from the center, 11 (31%) had LTP within 12 months. From the 12 tumors that had a unique positive specimen from the margin, 3 (25%) presented 12-month LTP. The difference was not statistically significant (p=1). A total of 17 tumors had more than one positive specimen, and 10/17 (56%) had LTP within 12 months. Tumors with more than one positive specimen were more likely to show evidence of LTP at 12 months (p=0.04).

Figure 7.

LTP-free survival stratified by immediate fluorescent assessment of the ablation zone.

Table 2.

Number of tumors at risk of local tumor progression after ablation

Number of Ablated Tumors at Risk	0 months	3 months	6 months	9 months	12 months
Negative for Tumor	91	91	85	75	71
Positive for Tumor	64	61	50	41	38

Complications

There were 2 arterial bleeding episodes; one of them corresponded to the area of post ablation biopsy. The second was more likely related to the ablation electrode position. Both were successfully embolized without further sequelae.

Discussion

This study validated prior observations that IFA can be used as a potentially intraprocedural, assay of ablated tumors [16]. It can provide proof of complete ablation of CLM or detect residual VT cells within 30 minutes from sampling. Nuclear labeling with Hoechst helps define neoplastic nuclei, and MT Red interrogates the mitochondrial activity and thus distinguishes viable from necrotic cells [22], resulting in a relatively high accuracy rate (94%) of ablated tissue assessment. Fluorescent stains are applied directly to the fresh tissue specimen followed by immediate microscopic evaluation that does not require further processing. Prior studies showed that the presence of Ki67-positive tumor cells adherent to electrodes after tumor ablation is associated with shorter LTP-free survival, disease specific survival and overall survival [14, 15]. Similarly, autofluorescence with glucose-6-phosphate diaphorase staining demonstrated that VT tissue adherent to ablation electrodes is an independent predictor of LTP after liver tumor RFA [12]. A common limitation of these studies is that tissue evaluation is performed after fixation and lengthy processing, thus not allowing for an immediate or timely assessment [12–14]. The potential for immediate assessment, while the patient is still in the IR suite could offer intraoperative feedback to the operator similarly to frozen section that is used to tailor tumor resections [23]. Such intraprocedural assessment and feedback can improve outcomes of tumor ablation, by allowing repeat ablation to those cases where residual VT is detected. When this is not feasible, the IFA will alert the treating physicians of the increased risk of treatment failure and LTP and the need for closer follow-up or adjuvant therapy. This methodology could improve oncologic outcomes of TA in the treatment of cancer and is probably a required step in the evolution of ablation as potential local cure with outcomes comparable to hepatectomy [24].

Reflectance spectroscopy has been used during RFA of CLM resulting in over 97% correlation with histopathological assessment and post-ablation CT imaging, but it could only distinguish ablated from non-ablated tissue [10]. The authors highlighted the value of the real-time feedback during ablation as a very important step in the effort to improve tumor ablation outcomes.

Fluorescent stains for immediate tissue assessment after ablation has been introduced with YO-PRO-1[11]. This fluorescent marker labels cells with compromised plasma membrane and has been used to assess apoptosis after RFA of CLM. Nevertheless, this method lacked specificity as similar levels of YO-PRO-1 signal were encountered in ablated and non-ablated animal liver specimens.

MT Red fluorescent stain assesses cell viability based on mitochondrial function addressing limitations of immediate tissue examination after ablation [16]. Specifically, previous work that studied resected specimens showed that tumor cells could undergo apoptosis several days after ablation [20]. This made immediate assessment of ablated tissue challenging as identification of morphologically intact tumor cells did not necessarily indicate untreated VT. It was clear that correct AZ assessment must evaluate the viability status of detected residual tumor cells by applying apoptotic and, more importantly, viability stains. Most of these stains require lengthy IHC processing precluding intraoperative assessments and feedback. The development of immediate/rapid assays and surrogate markers is critical for the evolution of ablation as a cancer therapy with local curative intent that could be offered instead of surgery [1, 2, 25].

The detection of VT in post-ablation tissue samples and the size of the ablation margin are the two most important technical factors associated with oncological outcomes [6]. Many studies have focused on optimizing the assessment of the ablation zone [26] and providing more accurate methods of measuring the ablation margin [27]. These methods have evolved from the side by side, visual comparison of pre- and post-ablation CT [1, 7] to 3D volumetric computation of the ablation margins, in order to address well-known limitations of the 2D assessment [27–31]. Leveraging the functional information provided by PET can also further improve AZ assessments [32–34]. Despite these improvements, patient, table and organ movement, as well as post-ablation inflammatory changes or local bleeding can complicate intraprocedural imaging assessments of the AZ and margins. More accurate AZ and margin measurements are thus performed after resolution of acute post-ablation changes, at least three weeks after ablation as recommended by the reporting standards for tumor ablation [21]. As a result, it is not rare to observe LTP even in tumors with AZ determined as technically successful with margins by imaging. Tissue examination is less vulnerable to operator variability and technical limitations, introducing the most objective surrogate of complete tumor ablation.

The added risk of biopsy is extremely small, especially in coagulated liver and certainly outweighs the risk of LTP that could be prevented by offering additional ablation on the spot, the next step through a new prospective study. Moreover, in order to eliminate this risk, a guiding needle sheath is used to allow the introduction of the core biopsy needle that often can be redirected to the center and MM of the AZ without repuncturing the liver surface.

This study was designed to validate the IFA of the AZ and it was not used to modify clinical ablation practice or decision pathways. No treatment decision was thus made based on the results of IFA, and ablation margins were not measured intraoperatively with image registration. The implementation of IFA in the treatment paradigm is currently assessed in combination with real-time 3D measurement of ablation margins in an open clinical trial currently enrolling participants (NCT01494324 blinded).

This study validated the fluorescent assessment (IFA) as an accurate real-time biomarker of complete tumor eradication or detection of residual viable tumor. In addition, the study demonstrated the prognostic value of IFA in LTP-free survival for patients undergoing TA of CLM. This development addresses a fundamental limitation of TA, proposes a potential shift in clinical ablation practice paradigm and could be possibly implemented in other image-guided locoregional therapies.

STROBE Statement—checklist of items that should be included in reports of observational studies

	Item No	Recommendation	✓ (N/A)
Title and abstract
	1	(a) Indicate the study’s design with a commonly used term in the title or the abstract	Y
		(b) Provide in the abstract an informative and balanced summary of what was done and what was found	Y
Introduction
Background/rationale	2	Explain the scientific background and rationale for the investigation being reported	Y
Objectives	3	State specific objectives, including any prespecified hypotheses	Y
Methods
Study design	4	Present key elements of study design early in the paper	Y
Setting	5	Describe the setting, locations, and relevant dates, including periods of recruitment, exposure, follow-up, and data collection	Y
Participants	6	(a) Cohort study—Give the eligibility criteria, and the sources and methods of selection of participants. Describe methods of follow-up Case-control study—Give the eligibility criteria, and the sources and methods of case ascertainment and control selection. Give the rationale for the choice of cases and controls Cross-sectional study—Give the eligibility criteria, and the sources and methods of selection of participants	Y
		(b) Cohort study—For matched studies, give matching criteria and number of exposed and unexposed Case-control study—For matched studies, give matching criteria and the number of controls per case	na
Variables	7	Clearly define all outcomes, exposures, predictors, potential confounders, and effect modifiers. Give diagnostic criteria, if applicable	Y
Data sources/ measurement	8*	For each variable of interest, give sources of data and details of methods of assessment (measurement). Describe comparability of assessment methods if there is more than one group	Y
Bias	9	Describe any efforts to address potential sources of bias	Y
Study size	10	Explain how the study size was arrived at	y
Quantitative variables	11	Explain how quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen and why	Y
Statistical methods	12	(a) Describe all statistical methods, including those used to control for confounding	Y
		(b) Describe any methods used to examine subgroups and interactions	Y
		(c) Explain how missing data were addressed	Y
		(d) Cohort study—If applicable, explain how loss to follow-up was addressed Case-control study—If applicable, explain how matching of cases and controls was addressed Cross-sectional study—If applicable, describe analytical methods taking account of sampling strategy	y
		(e) Describe any sensitivity analyses	Y
Results			✓ (N/A)
Participants	13*	(a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, examined for eligibility, confirmed eligible, included in the study, completing follow-up, and analysed	Y
		(b) Give reasons for non-participation at each stage	y
		(c) Use a flow diagram and include the figure number (preferably figure 1) or page number	y
Descriptive data	14*	(a) Give characteristics of study participants (eg demographic, clinical, social) and information on exposures and potential confounders	y
		(b) Indicate number of participants with missing data for each variable of interest	y
		(c) Cohort study—Summarise follow-up time (eg, average and total amount)	y
Outcome data	15*	Cohort study—Report numbers of outcome events or summary measures over time	y
		Case-control study—Report numbers in each exposure category, or summary measures of exposure	y
		Cross-sectional study—Report numbers of outcome events or summary measures	y
Main results	16	(a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and their precision (eg, 95% confidence interval). Make clear which confounders were adjusted for and why they were included
		(b) Report category boundaries when continuous variables were categorized	na
		(c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful time period	na
Other analyses	17	Report other analyses done—eg analyses of subgroups and interactions, and sensitivity analyses	na
Discussion
Key results	18	Summarise key results with reference to study objectives	y
Limitations	19	Discuss limitations of the study, taking into account sources of potential bias or imprecision. Discuss both direction and magnitude of any potential bias	y
Interpretation	20	Give a cautious overall interpretation of results considering objectives, limitations, multiplicity of analyses, results from similar studies, and other relevant evidence	y
Generalisability	21	Discuss the generalisability (external validity) of the study results	y
Other information
Funding	22	Give the source of funding and the role of the funders for the present study and, if applicable, for the original study on which the present article is based	y

^*Give information separately for cases and controls in case-control studies and, if applicable, for exposed and unexposed groups in cohort and cross-sectional studies.

Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is available at www.strobe-statement.org.

^*N/A stands for not applicable and may be a reasonable choice depending on the type of study performed

Research Highlights.

• Intraprocedural tissue immunofluorescent assay (IFA) was investigated in patients undergoing thermal ablation of colorectal liver metastases to detect residual viable tumor cells after ablation.

• For 155 ablated tumors in 99 patients, biopsies were performed of the ablation zone and of the margin, and evaluated for cellular morphology and mitochondrial viability.

• 27% of specimens were positive for viable tumor. Compared with subsequent standard histopathology, IFA was 94% accurate.

• Local tumor progression-free survival at 12 months was 59% for treated tumors positive for viable tumor cells vs 78% for IFA negative tumors.

• Intraprocedural IFA may provide actionable information regarding completeness of treatment.

Abbreviations

TA

Thermal Ablation

CLM

Colorectal Liver Metastases

LTP

Local Tumor Progression

AZ

Ablation Zone

IFA

Immediate Fluorescent Assessment

RFA

Radiofrequency ablation

MWA

Microwave ablation

H&E

Hematoxylin-eosin

IHC

Immunohistochemistry

MT Red

MitoTracker Red