Significant and Durable Clinical Response to Sorafenib and Radiation Therapy for a Patient With Stage IV Hepatocellular Carcinoma and LRRK2 Mutation

OVERVIEW

Sorafenib is an established therapy for unresectable/advanced hepatocellular carcinoma (HCC) on the basis of the SHARP (Sorafenib Hepatocellular Carcinoma Assessment Randomized Protocol) trial, in which sorafenib improved median overall survival compared with placebo (10.7 v 7.9 months).¹ We describe a patient who presented with stage IV HCC and developed a dramatic and prolonged response to sorafenib and radiotherapy. We discovered a previously unidentified LRRK2 mutation in the tumor that might account for the considerable sensitivity of this disease to radiation and sorafenib, and this hypothesis is supported by our preclinical investigations.

CASE REPORT

The patient was a 68-year-old man with history of noncirrhotic hepatitis B, hepatitis C, alcohol abuse, and intact liver function (Child-Pugh score A5) who presented with a 3-week history of intermittent epigastric pain. A computed tomography (CT) scan showed a 9.4 × 5.9 cm heterogeneous mass in the right hepatic lobe with substantial extrahepatic extension (Fig 1A, left panel) and an extrahepatic lesion associated with the transverse colon and mesentery that measured 3.8 × 2.8 cm (Fig 1A, right panel). The alpha fetoprotein (AFP) level was elevated to 57,400 ng/mL, and CT-guided biopsy of the large hepatic lesion revealed stage IV HCC. Three weeks after the baseline AFP was obtained, the patient started sorafenib 400 mg twice daily (Fig 1C). Repeat CT scans 7 weeks later showed notable disease progression with new, diffuse, multifocal hepatic lesions (Fig 1B). The two prior 9.4-cm and 3.8-cm masses also greatly increased in size and met RECIST criteria for progressive disease (Fig 1B). At 3.5 weeks after sorafenib was started, the AFP increased to 158,212 ng/mL but then decreased to 26,343 ng/mL by 8 weeks despite imaging that showed progression (Fig 1C). Because of radiographic concern for disease progression and invasion into the transverse colon, sorafenib was discontinued and radiation therapy was started 1 week after sorafenib was discontinued. The two large masses were irradiated to a dose of 40 Gy in 16 fractions using a conformal radiation plan (Fig 1D). Repeat CT scans 1 month later showed a large decrease in the size of not only the two radiated extrahepatic lesions but also the unirradiated hepatic lesions, and it was decided to restart the patient on maintenance sorafenib (200 mg twice daily) 6 weeks after radiation. Three months later, interval scans showed marked radiographic response of the two large abdominal masses and the smaller hepatic parenchymal lesions that received 10% to 20% of the radiation dose (Fig 1E). In addition, the AFP level normalized to 4.6 ng/mL. On the basis of this impressive response, sorafenib doses were held, and the patient underwent laparoscopic partial left hepatectomy of segment 4B with removal of additional primary liver and extrahepatic omental masses. Pathology from the liver resection showed a near-complete pathologic response with microscopic foci of HCC surrounded by necrosis and treatment-related changes. Pathology from the radiated extrahepatic abdominal masses showed no residual tumor. At 3 weeks after surgery, sorafenib was restarted at 200 mg twice daily but was discontinued after 2 more months because of sorafenib-related pulmonary toxicity. Fifty-nine months after sorafenib was first started, and 50 months after sorafenib was discontinued, there is no evidence of disease recurrence (Appendix Fig A1), and serial AFP levels have remained normalized (3.5 to 5.5 ng/mL). A timeline is shown in Appendix Figure A2. After this exceptional response, we performed next-generation sequencing of whole blood lymphocytes and tumor DNA from the diagnostic biopsy specimen (Table 1), which revealed mutations in LRRK2 (K1316N), LRP1B (V4250A, Y4256C), TP53 (V157insA, a nonframeshift insertion of three base pairs: CGC), and NOTCH3 (G501S). None of these variants are present in the dbSNP (Single Nucleotide Polymorphism Database), COSMIC (Catalogue of Somatic Mutations in Cancer), or Cancer Cell Line Encyclopedia databases.

FIG 1.

(A) Computed tomography (CT) of abdomen at presentation. (Left) 5.9 × 5.5 cm (axial) mass in the anterior right hepatic lobe with notable extrahepatic extension. (Right) 3.8 × 2.8 cm (axial) extrahepatic mass in the upper abdomen. There were no other lesions seen in the liver at this point. (B) CT of abdomen at 7 weeks after start of sorafenib. (Left) Interval appearance of diffuse intrahepatic lesions consistent with multifocal hepatocellular carcinoma (HCC). (Right) Interval progression of the two upper abdominal lesions. The lesion on the right increased from 5.5 × 5.9 cm to 7.2 × 7.2 cm on axial images. The medial lesion increased from 3.8 × 2.8 cm to 4.6 × 4.8 cm on axial images. (C) Alpha fetoprotein (AFP; ng/mL) values plotted versus time (months). (D) Radiation therapy (RT) dose distribution on CT simulation scan. The prescription dose was 40 Gy in 16 fractions at 2.5 Gy per fraction. Radiation was delivered with a three-field three-dimensional conformal plan using 6 MV photons. (E) CT of abdomen obtained 3 months after radiotherapy showed resolution of small hepatic parenchymal lesions and notable decrease in size of extrahepatic lesions. (Left) The right-sided lesion decreased to 3.1 × 2.5 cm on axial images. (Middle) The medial lesion decreased to 2.5 × 2.3 cm on axial images. (Right) Disappearance of intrahepatic lesions noted in Fig 1B.

TABLE 1.

Summary of Somatic Tumor Mutations Identified by Next-Generation Sequencing

LRRK2

We turned our attention to LRRK2, a gene that has been linked to the pathogenesis of Parkinson disease with known roles in the antioxidant response.^2,3 Ionizing radiation generates reactive oxygen species (ROS), which are critical for the ability of radiation to promote tumor cell kill through DNA damage. In addition, Coriat et al⁴ have demonstrated that sorafenib dose dependently induces the generation of ROS in tumor cells in vitro and in vivo. The degree of ROS generation correlates with serum levels of advanced oxidation protein products and clinical effectiveness in sorafenib-treated patients with HCC.⁴ Because ROS are critical in the response to both radiation and sorafenib, through promotion of DNA damage, we hypothesized that this mutation in LRRK2 could alter the response to both therapies. We obtained a full-length human cDNA for LRRK2 and created the identified mutation (K1316N) using site-directed mutagenesis. We observed adequate expression of transfected wild-type (WT) and K1316N leucine-rich repeat kinase 2 (LRRK2) protein in multiple cell lines (Appendix Fig A3 for RNA and protein; primer sequences inTable A1). Then, we tested the effects of WT and K1316N on proliferation in the HCC cell lines Hep3B and HepG2 and found minimal effect (Appendix Fig A4). We then tested the effects of WT and K1316N on sorafenib sensitivity and found that the LRRK2 K1316N mutant rendered both HepG2 and Hep3B cells more sensitive to sorafenib, whereas WT LRRK2 had minimal effect on sorafenib sensitivity (Figs 2A and 2B). We also tested genetic silencing of LRRK2 with small interfering RNA (siRNA) and found that LRRK2 depletion likewise sensitized HCC cells to sorafenib (Figs 2C and 2D). In addition, we found that expression of the LRRK2 K1316N mutation in both HeLa and HepG2 cells resulted in notable radiation sensitization in radiation clonogenic assays, whereas WT LRRK2 had no radiosensitizing effects (Figs 2E and 2F). Because of the previously identified role of LRRK2 in ROS, we assessed whether WT LRRK2 or the K1316N mutant altered levels of ROS in cancer cells. Expression of LRRK2 K1316N in Hep3B, HepG2, and HeLa cells markedly increased superoxide (O₂⁻) levels compared with empty vector cells, and the increase of superoxide was reversed with the antioxidant tiron (Figs 3A, 3B, and 3C). Similarly, we found that LRRK2 K1316N greatly increased hydrogen peroxide (H₂O₂) levels, another known mediator of ROS damage, and this increase was reversed by treatment of cells with ebselen, an H₂O₂ scavenger (Figs 3D, 3E, and 3F). Interestingly, in Hep3B and HeLa cells, WT LRRK2 also increased superoxide and H₂O₂ levels, but these effects were more minimal/absent in HepG2 cells. Finally, the addition of the superoxide scavenger tiron greatly attenuated the radiation sensitivity induced by LRRK2 K1316N in radiation clonogenic assays (Figs 4A and 4B).

FIG 2.

LRRK2 K1316N mutation sensitizes hepatocellular carcinoma (HCC) cells to sorafenib and radiation treatment. (A, B) Overexpression of LRRK2 K1316N significantly increases sorafenib sensitivity on HCC cells. (C, D) Genetic silencing of LRRK2 with siRNA sensitizes HCC cells to sorafenib treatment. (E, F) Overexpression of LRRK2 K1316N sensitizes HepG2 and HeLa cells to radiation treatment. Sorafenib sensitivity on HCC cells was determined by alamarBlue assay (ThermoFisher, Waltham, MA) 72 hours after transfection, whereas radiation sensitivity was determined by clonogenic assay. (*) P < .05; (†) P < .001. Note that the human cervical cancer cell HeLa was used for the clonogenic assay, because the Hep3B cell line does not form colonies. siCtr, control (scrambled) siRNA; siLRRK2, LRRK2 siRNA; EMP, empty vector; WT, wild type.

FIG 3.

LRRK2 K1316N mutation enhances reactive oxygen species (ROS) activity in hepatocellular carcinoma (HCC) and HeLa cells. (A,B,C) Overexpression of LRRK2 K1316N significantly increased superoxide levels compared with empty vector cells. The antioxidant tiron was used to determine the specificity of ET probes. (D,E,F) Expression of LRRK2 K1316N enhanced hydrogen peroxide levels. Ebselen was used to determine the specificity of dihydrofluorescein (HFLUOR; Sigma-Aldrich, St Louis, MO) probes. (*) Significant difference from corresponding tiron or ebselen treatment group; (†) significant difference from empty vector; (‡) significant difference from LRRK2 wild type (WT). P < .05.

FIG 4.

Antioxidant tiron attenuates LRRK2 K1316N radiosensitization in HepG2 and HeLa cells. HepG2 and HeLa cells transfected with wild-type (WT) LRRK2 and LRRK2 K1316N were preincubated with DMSO or tiron (1 mM) and then exposed to 4 Gy of radiation. Cells were cultured for approximately 2 weeks for colony formation. Tiron significantly increased colony numbers in (A) HepG2 and (B) HeLa cells that overexpressed LRRK2 K1316N. (*) P < .001. IR, irradiation.

DISCUSSION

In this case of a patient with an exceptional response to sorafenib and radiation, we identified a novel LRRK2 K1316N mutation that induced sorafenib and radiation sensitivity in various HCC cell lines, consistent with our clinical observations. In addition, we found that this LRRK2 mutation promoted increased ROS, which perhaps accounted for a mechanism of increased therapeutic sensitivity. LRRK-2 enzymatically activating mutations have been identified as the most frequent cause of inherited autosomal dominant Parkinson disease but have also been found in the more common sporadic form of Parkinson disease.² The most common activating mutation, G2019S, causes increased production of ROS, which could lead to neuronal degeneration.³ In addition, another study of LRRK2 in neurodegeneration showed that LRRK2 kinase directly phosphorylates the tumor suppressor protein p53, which leads to enhanced nuclear retention, increased p21 (WAF1/CIP1) expression, and subsequent tumor suppressor activity by promoting cell cycle/growth arrest and/or apoptosis.⁵ In relation to cancer, perhaps the presence of a hyperactive LRRK2 mutant could promote increased p53 activity and thereby lead to tumor cell death and apoptosis in the setting of additional oxidative stress, radiation, and/or sorafenib. Thus, we hypothesize that this LRRK2 mutation increases LRRK2 kinase activity and enhances sorafenib and radiation-induced apoptotic cell death signaling via increased p53 and p21 activity.

Our in vitro data suggest that this LRRK2 K1316N mutation increases ROS, like other canonical LRRK2 mutations observed in Parkinson disease. Thus, another potential hypothesis to explain the radiosensitivity of disease in this patient is that increased LRRK2 kinase activity in LRRK2 mutants may render cells more susceptible to oxidative stress imparted by radiotherapy through creation of ROS, such as oxygen free radicals.⁶ Multiple publications show that LRRK2 mutations induce synergistic or additive cell killing in the presence of oxidative stress.^7,8 Taken together, the presence of increased oxidative stress imparted by an activating LRRK2 mutation could prime tumor cells to increased radiation-mediated cell death.

The near-complete pathologic response and more than 4-year disease-free interval since completion of all therapy are remarkable results from a low to moderate dose of radiation and a drug that rarely induces complete response (0% in the SHARP trial). It is also interesting to note the dramatic radiographic responses seen in lesions that were not directly radiated, which suggests an abscopal effect from the combination of sorafenib and radiotherapy. Certainly, we cannot overlook that the dramatic responses observed may have been due in part to the development of an antitumor immune response, which has been previously attributed to both sorafenib and radiation.^9,10 Indeed, the radiologic progression initially noted after sorafenib started may be related to pseudoprogression and development of an early immune response or apoptosis/necrosis of the tumor, despite the observation of the most dramatic clinical and radiographic response with postradiation sorafenib.

In conclusion, these data support a role for a novel, previously unidentified LRRK2 mutation that promotes sorafenib and radiation sensitivity through an ROS-dependent mechanism. In addition, the combination of radiation and sorafenib also may have induced an antitumor immune response. Additional research is needed to more directly establish the role of this mutation in the promotion of sensitivity to radiation, sorafenib, and/or an antitumor immune response.

Appendix

FIG A1.

Computed tomography of the abdomen obtained 2 years after surgery. Stable post-operative changes are seen, and there is no evidence of disease.

FIG A2.

Timeline of key diagnostic, therapeutic, and radiographic events. AFP, alpha fetoprotein; CT, computed tomography; HCC, hepatocellular carcinoma.

FIG A3.

Quantitative polymerase chain reaction (qPCR) analysis (A) and Western blot (B) confirmed overexpression of wild-type (WT) LRRK2 and LRRK2-K1316N (mutant) on four cell lines 72 hours after transfection. Empty vector plasmid (EMP) was used as control.

FIG A4.

Overexpression of wild type (WT) or K1316N mutant LRRK2 showed minimal effects on (A) HepG2 and (B) Hep3B cell proliferation, determined by alamarBlue assay at 72 hours after transient transfection.

TABLE A1.

Primer Sequences for qPCR Analysis

AUTHOR CONTRIBUTIONS

Conception and design: Linlin Yang, Anne Noonan, Evan Wuthrick, Carl Schmidt, Terence Williams

Collection and assembly of data: Linlin Yang, Patrick Wald, Sameek Roychowdhury, Anne M. Noonan, Li Zuo, Chia-Chen Chuang, Julie Reeser, Evan Wuthrick, Terence Williams

Provision of study material or patients: Sameek Roychowdhury, Anne M. Noonan

Data analysis and interpretation: Linlin Yang, Patrick Wald, Sameek Roychowdhury, Li Zuo, Chia-Chen Chuang, Evan Wuthrick, Carl Schmidt, Terence Williams

Manuscript writing: All authors

Final approval of manuscript: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. 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.

Patrick Wald

Employment: Riverside Radiation Oncology

Sameek Roychowdhury

Stock and Other Ownership Interests: Johnson & Johnson (I)

Consulting or Advisory Role: Incyte, AbbVie, QED Therapeutics

Research Funding: Takeda, Ignyte

Anne M. Noonan

Consulting or Advisory Role: Helsinn Healthcare

No other potential conflicts of interest were reported.