Empowering Clinical Decision Making in Oligometastatic Colorectal Cancer: The Potential Role of Drug Screening of Patient-Derived Organoids

The definition of oligometastatic colorectal cancer (CRC) identifies a peculiar subpopulation of patients characterized by a limited metastatic spread of disease. Oligometastatic disease is defined as the involvement of up to two or occasionally three sites with five or sometimes more metastases that for their anatomic localization is amenable to local ablative therapies, thus rendering the patient free of disease. Thus, this subgroup is wide and has a significant cohort of patients with CRC. Among patients with oligometastatic CRC, those with liver-limited disease represent a more refined subset and should always be discussed in multidisciplinary teams since they appear to more likely benefit from multimodal approaches with curative intent. A perioperative systemic treatment integrated with surgical liver metastasectomy should be regarded as the best multimodal approach. However, the best drug regimen to be adopted for this subset of patients with CRC is still debatable and should be tailored case-by-case.


INTRODUCTION
The definition of oligometastatic colorectal cancer (CRC) identifies a peculiar subpopulation of patients characterized by a limited metastatic spread of disease. 1 Oligometastatic disease is defined as the involvement of up to two or occasionally three sites with five or sometimes more metastases that for their anatomic localization is amenable to local ablative therapies, thus rendering the patient free of disease. [1][2][3] Thus, this subgroup is wide and has a significant cohort of patients with CRC. Among patients with oligometastatic CRC, those with liver-limited disease represent a more refined subset and should always be discussed in multidisciplinary teams since they appear to more likely benefit from multimodal approaches with curative intent. 2,[4][5][6] A perioperative systemic treatment integrated with surgical liver metastasectomy should be regarded as the best multimodal approach. 2,7 However, the best drug regimen to be adopted for this subset of patients with CRC is still debatable and should be tailored case-by-case.
Considering left-sided microsatellite stable, RAS and BRAF wild-type CRC, doublet cytotoxic regimens (FOLFIRI or FOLFOX) plus an anti-epidermal growth factor receptor (EGFR) drug can represent the best option on the basis of significant response rate (RR). [8][9][10] However, triplet cytotoxic regimens FOLFOXIRI plus bevacizumab or anti-EGFR drug demonstrate an impressive RR up to 87%, which are increasingly regarded as potential novel neoadjuvant standard strategies. 5,11 However, severe treatment-related toxicities have been reported in up to 80% of patients. 11,12 Hence, both doublet or triplet combinations and anti-EGFR or antivascular endothelial growth factor are feasible options in this subset of patients.
The generation of preclinical models such as patientderived organoids (PDOs), recapitulating patient tumor histology and genetics, is emerging as a tool to predict treatment efficacy in oncology. 13,14 Although genomics has already improved treatment choice in patients with CRC, especially for those carrying RAS/ BRAF wild type, coclinical trials are becoming more and more important to directly test different treatment options in patient-derived tumors. 15 Here, we present a proof-of-concept case report about the potential role of drug sensitivity testing in PDOs in the clinical decision making of oligometastatic CRC.

TRANSLATIONAL REPORT
In July 2018, a 56-year-old healthy man underwent a screening colonoscopy and was diagnosed with a leftsided colonic mass demonstrating a moderately differentiated (G2) adenocarcinoma. Afterward, a computed tomography (CT) scan revealed two metastatic lesions in liver segments VIII and V (Fig 1A). Baseline carcinoembryonic antigen (CEA) was 15.5 ng/mL, and cancer antigen (CA) 19.9 was 55 U/mL. Following multidisciplinary discussion and on the basis of standard molecular biomarker assessment, demonstrating microsatellite stability, no mutations in RAS and BRAF, and no ERBB2 amplification, the patient received four cycles of neoadjuvant FOLFOX plus panitumumab (Fig 2). Treatment was complicated by grade 3 afebrile neutropenia requiring peg-filgrastim support. No other adverse events occurred.
CT scan reassessment demonstrated partial response (PR) (Figs 1B and 1E), and thus, following a new multidisciplinary discussion, in November 2018, the patient underwent R0 liver metastasectomy and left emicolectomy. The pathology report described an ypT3 ypN2b ypM1aG2 adenocarcinoma. At this time, after signing informed consent and the enrollment in the AlfaOmega observational trial (NCT04120935), a fresh tumor sample from the metastatic liver lesion in segment VIII was used for PDOs generation. Once established (Data Supplement), these tumoroids underwent histologic analysis, confirming the correspondence and similarity between the PDOs and the metastatic tissue of origin (Fig 2). The next-generation sequencing performed by the FoundationOne CDx panel on primary left colon lesion, as per GO40782 clinical trial screening (NCT02568267), revealed no mutations discriminating for sensitivity to cytotoxic agents. 2,6, 16 We also performed whole-exome sequencing analysis on PDOs, confirming the presence of patient-specific trunk alterations, whereas no alterations conferring resistance to anti-EGFR drugs were observed (Fig 2).
Following surgical resection, as per clinical practice, the patient received further eight cycles of systemic treatment with FOLFOX regimen supported with peg-filgrastim prophylaxis, ending in April 2019. The patient tolerated the postoperative treatment well without relevant side effects.
Despite this multimodal approach, CT and positron emission tomography scan in October 2019 showed disease relapse in liver segments VIII, VII, and V ( Fig 1C). Considering this evidence, the case was discussed collectively and we also took into consideration the results of a drug sensitivity testing performed on the PDOs, which, evaluating the most effective agents in this clinical context, showed complete resistance to oxaliplatin (OXA) but rather a prominent sensitivity to fluorouracil 5-FU SN-38 (the active metabolite of irinotecan), panitumumab, and cetuximab (Fig 3 and Appendix Fig A1). Primarily based on clinical standard and corroborated by PDOs drug sensitivity testing, we avoided OXA-based treatment and the patient was treated with six cycles of FOLFIRI and panitumumab and peg-filgrastim prophylaxis. This schedule was welltolerated, and the following CT scan again demonstrated a PR (Figs 1D and 1E). Given this result, in March 2020, the patient underwent multiple liver metastasectomy confirming metastatic adenocarcinoma. Thereafter, the patient received six cycles of postoperative 5-FU, again welltolerated. In July 2020, the CT scan demonstrated no evidence of disease and CEA and CA19.9 were normal. The following CT scan in October 2020 confirmed no relapse with normal CEA and CA19.9. The patient has consented to the submission and the publication of the case report.

DISCUSSION
Our groups have jointly focused their efforts on the development of preclinical models with the aim of tailoring treatment for patients with metastatic CRC on the basis of the results obtained from the pharmacologic screening on patients' avatars. [17][18][19][20][21][22][23][24] By taking advantage of patientderived xenografts, 19 we previously identified optimal therapeutic regimens for CRC druggable targets such as ERBB2 amplification. [25][26][27] Recently, the establishment of organoids directly obtained from surgical procedure or After six cycles of FOLFIRI and panitumumab, the CT scan performed in January 2020 showed a decrease in lesions' size: segment V decrease by 50% (from 12 to 6 mm), segments V-VIII by 37% (from 19 to 12 mm), and segment VII by 33% (from 18 to 12 mm). All lesions were evaluated as per RECIST1.1 criteria. (E) Bar graph summarizes the best RECIST1.1 response to both first-line treatment with FOLFOX plus panitumumab (-45%) and second-line treatment with FOLFIRI plus panitumumab (-39%). CT, computed tomography; Pmab, panitumumab.
tumor biopsy has accelerated the possibility to test drugs on patients' avatars, 21,23,28 thus offering a valid approach to test drug sensitivity on the bench, in parallel to patient care.
Here, we report the case of an patient with oligometastatic CRC whose drug screening on PDOs resembles both resistance and sensitivity to main cytotoxic agents. Most notably, PDOs drug testing was performed in parallel with clinical care. This translational paralleling is usually hampered by the time required to establish PDOs and test drug activity in the continuum of care of CRC (Fig 2). However, the expected recurrence-free survival following liver resection in oligometastatic CRC offers a timely opportunity to translate the PDOs drug sensitivity testing results into clinical decision making at relapse. Here, we were able to identify a medical perioperative regimen with the higher chance to obtain tumor shrinkage, aiming to offer a second potentially curative resection to the patient. 2  metastasectomy to improve the chance of cure, sparing side effects of other intensive cytotoxic regimens. 11,[29][30][31] Here, we limited our screening to approved drugs, but elsewhere, we already proposed rationale combinations including unapproved therapeutics on the basis of the results from drug screening on PDOs. 17  This case report reinforces other recent publications in which PDOs were suggested as a potential platform to identify the best treatment for each patient. 17 In addition, we envision the subset of patients with oligometastatic CRC as the optimal to bypass the limitation given by time required to PDOs establishment. Prospective translational trials are needed to verify the feasibility of this translational approach in peculiar clinical settings (Table 1).
Treatment implementation driven by PDOs drug sensitivity testing has intrinsic limitations potentially hampering its applicability in the daily clinical practice. First, limiting factors such as tumor heterogeneity leading to potential sampling biases might represent relevant barriers toward translational applicability. Thus, we hypothesize that PDOs testing might be more helpful in patients with a low tumor burden, because of intrapatient heterogeneity that is expected to be lower in these patients. 34,35 Indeed, diffused metastatic disease is more likely to be widely heterogeneous and patients do not achieve the same results observed on PDOs. 21,36,37 Second, PDOs drug sensitivity testing constitutes a relatively expensive method to choose a systemic cytotoxic regimen. However, we reasoned that this might be justifiable to maximize the chance of cure and improve the quality of life. Finally, in hospitals where a close connection with a preclinical laboratory is not a reality, PDO-driven approach might be difficult to be set up.
In conclusion, we provide a translational report demonstrating that PDOs drug sensitivity testing is feasible and might represent a valid tool to predict and improve the choice for the best perioperative personalized approach in patients with oligometastatic CRC. We suggest indeed that this setting, because of the time intervals between surgery and resumption of medical treatment, would offer the best choice to parallel patients' care, allowing timely modulation of the treatment strategy. These advantages should be supported by optimization of protocol standardization and balance against costs related to organoid maintenance.

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
The following represents disclosure information provided by the authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs. org/po/author-center. Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).