Hepatocellular Carcinoma Over the Past Decade

Tanios S. Bekaii-Saab, MD, FACP

In conjunction with the 10-year anniversary of Targeted Oncology™, we are highlighting the most impactful advancements concerning hepatocellular carcinoma (HCC) over the previous decade.

Tanios S. Bekaii-Saab, MD, FACP, medical oncologist, medical director, Cancer Clinical Research Office, vice chair and section chief, Medical Oncology, Department of Internal Medicine, Mayo Clinic; Richard S. Finn, MD, professor of medicine, Division of Hematology/Oncology, director, Signal Transduction and Therapeutics Program at UCLA Jonsson Comprehensive Cancer Center, David Geffen School of Medicine at UCLA; and William P. Harris, MD, associate professor in the Division of Oncology at the University of Washington School of Medicine in Seattle offered their thoughts and experience concerning the highlights over their previous 10 years of research and clinical treatment of this disease.

Etiology/Pathogenesis

Improved understanding of the pathogenesis of a disease and the development of effective treatments generally results in a decrease in the incidence and mortality rates of cancer. Despite advancements in the field of HCC, incidence and mortality rates are continuing to increase globally.¹ Within the United States specifically, incidence rates increased over the first half of the decade but have begun to slightly decrease while mortality rates have more or less remained steady (FIGURE).²

The greatest historical risk factors for developing HCC have been hepatitis B virus (HBV) and hepatitis C virus (HCV).³ Since 2014, several antiviral options for HCV have been approved, resulting in higher rates of sustained virological response (SVR).⁴ SVR has been associated with a considerable decreased risk of HCC, although the risk is still present for patients with HCV and cirrhosis. Proper vaccination programs and availability of HBV vaccines have greatly decreased incidence in countries such as the United States, but HBV remains a global concern, especially in Asia.⁵

With the widespread use of vaccines for HBV and the availability of antivirals for HCV, countries such as the United States could experience a drop in the incidence of HCC. “The primary driver was viral hepatitis; however, the pathogenesis of HCC is shifting and continues to shift,” stated Saab, “more and more, due to the incidence of increased obesity and non-alcoholic fatty liver disease [NAFLD], NASH [non-alcoholic steatohepatitis] is becoming the driver.” NASH and NAFLD are significantly contributing to the current number of HCC cases and predicted to result in continually increasing incidence rates.⁶

NASH and NAFLD are related to obesity and type 2 diabetes mellitus, which can be driven by both environmental and genetic factors.6 The projected prevalence of NAFLD by 2030 is 33.5%, inevitably resulting in increased mortality rates.⁷ Harris hopes we can prevent future HCC by targeting at-risk patients for hepatic disease. “We need to do a better job recognizing high-risk patients. Treatment of hepatitis, obesity, and diabetes can decrease the risk of NASH and HCC.” Finn also sees NASH as the next hurdle to overcome: “Incidence from hepatitis C is expected to decline, but NASH is a very potent risk factor for many patients and is becoming the next battle in the primary space.”

Diagnosis/Surveillance—The Role of Imaging

Richard Finn, MD

Imaging provides a non-invasive approach to diagnosis, staging, and follow-up, and in recent years, the approach to imaging in HCC has shifted from a qualitative to a more quantitative approach.⁸ The quantitative approach is an attempt to use imaging parameters such as spatial differences in brightness to help create a reproducible signature that accurately represents the given pathophysiology. In order to assist in standardizing the approach to HCC diagnosis and improve communication between clinicians, the Liver Imaging Reporting and Data System (LI-RADS) was developed. LI-RADS continues to be updated and integrate various modalities (CT, MRI, contrast-enhancing ultrasound) and conforms to the American Association for the Study of Liver Diseases (AASLD) disease recommendations.^9,10 “The main changes for diagnostic imaging has really been in the standardization of reporting with the LI-RADS system. This is becoming a standardized mechanism to determine which lesions appear definitive for liver cancer radiographically and has begun to incorporate parameters on treatment response,” Harris said.

Although the standardization of imaging technique and assessment through LI-RADS certainly increases the likelihood of a proper diagnosis, it still lacks the desired sensitivity and specificity. Up to 38% of LI-RADS 3 classifications (intermediate probability of malignancy) and 13% of LI-RADS 2 classifications (probably benign) are confirmed as HCC by pathological diagnosis.¹¹

Researchers continue to work toward improved MRI and CT techniques to better assess preoperative parameters, such as microvascular invasion along with liver status and portal hypertension. Numerous approaches have also been developed and are continuing to be developed to aide with prognosis and properly assess treatment response. Currently there remains no standardized approach, and no imaging study has been able to accurately determine histologic subtypes of HCC.⁸

Finn appreciates the advancements but points out more education is needed for oncologists in the community, “If you’re in the liver cancer space you understand the LI-RADS, but many in the community won’t. There’s still a lot of education needed to understand how to follow patients with liver cancer, how to assess response.”

Saab admits the screening process is important but requires further optimization and has concerns about access issues for at-risk population groups within the United States. “Current imaging screening is still important, and picks up patients in earlier stages, but we’re not where we want to be yet. There is also an access issue in the United States. We should be focusing on patients with higher risk, but we see a lot of liver disease in patients with decreased access, such as Hispanics and Native Americans.”

Locoregional Treatment

Transarterial chemoembolization (TACE) remains the standard-of-care treatment for patients with intermediate-stage HCC.³ However, the recent development of transarterial radioembolization (TARE) could potentially change that in the future.¹² TARE spares healthy liver parenchyma by delivering high-dose β radiation to the capillary bed of the tumor. A catheter is used to administer glass or resin microspheres embedded with yttrium-90–emitting β radiation. Current data suggest TARE is an effective option for liver-limited, unresectable disease.^13,14 Data from a meta-analysis suggest that TARE is not as effective as TACE or TACE with drug-eluting beads (DEB) in terms of overall survival (OS), but that TARE produces significantly fewer complications.15 Further studies are needed before TARE could possibly replace TACE, but many clinicians already prefer TARE to TACE for its tolerability.

William P. Harris, MD

Harris mentioned that while there have not been many additional locoregional approaches, the available modalities are continuing to improve and become more precise. “Local interventions have developed over the past 10 years with the understanding that we want to be more selective in our treatments while sparing healthy liver parenchyma. Localized interventions will continue to be honed. Promising outcomes are currently achieved for treating single tumors up to 8 cm in size with or without focal vascular invasion with high-dose ablative radioembolization administered selectively. We see response rates pushing 90% with durable responses.”

Systemic Treatments

The greatest advancements over the previous decade in HCC have been the considerable number of systemic therapeutic approvals (TIMELINE).

The multikinase inhibitor (TKI) sorafenib (Nexavar) was approved as first-line therapy for HCC in 2008 based off the results of the SHARP trial (NCT00105443).16 Regarding sorafenib, Finn commented, “Sorafenib improved survival but didn’t have a large impact on objective response rate, continuing our reliance on interventional techniques.” Saab added, “It’s an incredibly tough drug to use, but it was a welcome option as there was nothing else available.” Many thought sorafenib would quickly lead to further drug developments and approvals, but there was a drought of negative trials for the next decade. The next drug approval wasn’t until 2017, but that led to a flood of new approved drugs for HCC.

Regorafenib

The multikinase inhibitor regorafenib (Stivarga) targets vascular endothelial growth factor receptor (VEGFR) 1-3, among other kinases, and was the first second-line therapy approved.¹⁷ It was labeled for the treatment of patients who previously received sorafenib. Data from the RESORCE trial (NCT01774344) was modest with regorafenib producing a median OS of 10.6 moths vs placebo at 7.8 months (HR, 0.63; 95% CI, 0.50-0.79; P < .0001), median progression-free survival (PFS) of 3.1 months vs 1.5 months (HR, 0.46; 95% CI, 0.37-0.56; P < .001), and an objective response rate (ORR) of 11% vs 4% (P = .0047).¹⁷

Lenvatinib

In 2018 the TKI lenvatinib (Lenvima) was approved for the first-line treatment of patients with unresectable HCC. Approval was based on the phase 3 non-inferiority REFLECT trial (NCT01761266), which demonstrated that lenvatinib was noninferior to sorafenib in patients with advanced, untreated HCC.^3,18 Lenvatinib showed an improved median OS over sorafenib (13.6 vs 12.3 months; HR, 0.92; 95% CI, 0.79-1.06). It also achieved a significantly improved PFS and ORR.¹⁸ However, lenvatinib’s rate of adverse events was similar to sorafenib with grade 3 or higher drug-related adverse events in almost half of all treated patients. Saab highlighted, “The noninferiority study doesn’t technically show superiority of lenvatinib. Lenvatinib won in some ways because it was a little more manageable than sorafenib and became the preferred agent for many.”

Pembrolizumab

Pembrolizumab (Keytruda), a PD-1 inhibitor, was approved based on a durable ORR of 17% and median OS of 12.9 months in patients previously treated with sorafenib from the phase 2 KEYNOTE-224 trial (NCT02702414).¹⁹ However, the follow-up phase 3 study KEYNOTE-240 (NCT02702401) missed the pre-specified P value required for statistical significance with a median OS of 13.9 months for pembrolizumab vs 10.6 months for placebo (P = .02) and a median PFS of 3.0 months vs 2.8 months for pembrolizumab and placebo, respectively (P = .0186).²⁰ This placed uncertainty on the standing of the indication.

In 2021, the FDA’s Oncologic Drugs Advisory Committee discussed the pembrolizumab HCC indication, among others, and unanimously voted in favor of continued approval in the second-line setting.²¹

Data presented at the Gastrointestinal Cancers Conference 2022 from the phase 3 KEYNOTE-394 trial (NCT03062358) indicate a significant improvement in OS for pembrolizumab compared with placebo, both plus best supportive care, for patients with previously treated advanced HCC. The risk of death was decreased by 21% with pembrolizumab (HR, 0.79; 95% CI, 0.63-0.99; P = .0180) with a median OS of 14.6 months compared with 13.0 months for placebo.²² This will likely keep pembrolizumab as a second-line monotherapy option despite the KEYNOTE-240 results.

Cabozantinib

Cabozantinib (Cabometyx) is another multikinase inhibitor that received approval based on the CELESTIAL trial (NCT01908426) revealing a median OS of 10.2 months for cabozantinib compared with 8 months with placebo (HR, 0.76; 95% CI, 0.63-0.92; P = .0049). The median PFS was 5.2 months with cabozantinib and 1.9 months with placebo (HR, 0.44; 95% CI, 0.36-0.52; P < .001). ORRs were less than 10% in both treatment arms.23 National Comprehensive Cancer Network (NCCN) guidelines currently recommend cabozantinib as a category 1 option for patients with previously treated Child-Pugh class A disease.²⁴

Ramucirumab

Ramucirumab (Cyramza) is a VEGFR antagonist and is the only biomarker-guided therapy for HCC, requiring a baseline α-fetoprotein (AFP) level of 400 ng/mL or more.²⁵ In the phase 3 REACH-2 trial (NCT02435433), ramucirumab produced an OS of 8.5 months vs 7.3 months for placebo (HR, 0.710; 95% CI, 0.53-0.949; P = .0199) in previously treated patients with HCC.

Nivolumab and Ipilimumab

Nivolumab (Opdivo) was initially approved as a second-line monotherapy treatment in 2017 based on results of a portion of the phase 1/2 CheckMate 040 study (NCT01658878).²⁶ In the trial, nivolumab monotherapy induced an ORR of 20%. However, the confirmatory phase 3 CheckMate 459 trial (NCT02576509) did not show a statistically significant improvement in OS for nivolumab over sorafenib (median OS, 16.4 vs 14.7 months, respectively; HR, 0.85; 95% CI, 0.72-1.02; P = .075), resulting in the withdrawal of the indication.^27,28

The combination of nivolumab (anti–PD-1 inhibitor) and ipilimumab (CTLA-4 monoclonal antibody; Yervoy) was also explored in the CheckMate 040 trial. An ORR of 32% and a median OS of 22.8 months was achieved by the nivolumab plus ipilimumab group.²⁹ Based on these results, the FDA granted accelerated approval to the combination despite immune-related toxicities that required systemic corticosteroid use in 51% of cases. Phase 3 trials are ongoing comparing this combination with sorafenib or lenvatinib.³

Atezolizumab and Bevacizumab

The most important approval and overall advancement in the treatment of HCC was the approval of the anti–PD-L1 antibody atezolizumab (Tecentriq) with the anti-VEGF antibody bevacizumab (Avastin). This combination is the first regimen to significantly improve OS vs sorafenib in the frontline setting.^30,31 IMbrave150 (NCT03434379) was an open-label, randomized phase 3 trial comparing the combination of atezolizumab and bevacizumab with sorafenib as first-line therapy for advanced HCC. Updated analysis indicated an OS of 19.2 months for the combination arm and 13.4 months for sorafenib as monotherapy (HR, 0.66; 95% CI, 0.52-0.85; P = .0009) along with a median PFS of 6.8 months for the combination arm vs 4.3 months for sorafenib (HR, 0.59; 95% CI, 0.47-0.76; P < .001). The ORR was 30% in the combination arm with a duration of response of 18.1 months compared with 11% for sorafenib with a duration of response of 16.3 months. Not only were efficacy results greater, but patients also reported higher quality of life and tolerability.^30,31

This approval shifted the preferred first-line treatment from sorafenib monotherapy to atezolizumab with bevacizumab. “Atezolizumab and bevacizumab showed significant improvement compared to sorafenib along with improved quality of life. This truly saved immuno-oncology therapy by becoming the standard for patients with HCC,” stated Saab.

Biomarkers

Biomarker data to help guide treatment is still lacking in HCC. Elevated serum AFP is a marker of poor prognosis across HCC and is predictive of response with the use of ramucirumab.³² Decreasing AFP levels have been related to increased PFS and OS while an increase in AFP levels has been associated with tumor progression.³³ Due to the complexity of HCC, characterizing the disease based on a single biomarker is difficult and AFP carries low sensitivity and specificity as a means of surveillance.³³

Harris pointed out there are trials underway such as the GALAD trial to evaluate other biomarkers for HCC early detection.³⁴ However, larger phase 3 trials are needed before they can be relied upon.

Looking Ahead

Harris, Finn, and Saab all agree that further studies into second-line therapy after immuno-oncology therapeutics, biomarkers for diagnosis and treatment selection, and personalized therapies for patients require more investigation in the years to come. Several studies have their attention including the phase 3 LEAP-002 study (NCT03713593) evaluating lenvatinib plus pembrolizumab for first-line treatment of patients with advanced HCC.³⁵ Recent outcomes from the COSMIC-312 trial (NCT03755791) evaluating the combination of cabozantinib and atezolizumab vs sorafenib for first-line therapy are encouraging, though not as robust as results observed with atezolizumab with bevacizumab.³⁶ The HIMALAYA trial (NCT03298451) recently found that the anti–PD-1 antibody durvalumab (Imfinzi) is noninferior alone compared with sorafenib and produces a meaningful OS benefit when given with a high priming dose of tremelimumab (an anti–CTLA-4 antibody) in patients with unresectable HCC as frontline therapy.³⁷

These preliminary results and study designs are encouraging, yet the significance of their impact is questionable. These new treatments will provide options, but the data are similar or even less robust compared with the options already available.

“In my opinion, we need new strategies outside of exploring our current options and how to sequence them,” stated Harris. Saab agrees, “In HCC right now it’s primarily assessing what we have, but I do not anticipate much change in this coming year. How can we enhance the activity of the already active regimens? Adding other components or moving into CAR [chimeric antigen receptor] T cells and other immune-based therapies seem to make sense as possibilities.”

Finn concluded, “We’ve seen significant gains in overall survival and that’s very exciting coming from a decade where we were improving survival but with stable disease. Now we’re actually seeing tumors shrink.”

_References

_{1. McGlynn KA, Petrick JL, El-Serag HB. Epidemiology of Hepatocellular Carcinoma. Hepatology. 2021;73(Suppl 1):4-13. doi:10.1002/hep.31288}

_{2. Liver and Intrahepatic Bile Duct: Recent Trends in SEER Incidence (2000-2018) and U.S. Mortality (2000-2019) Rates. National Cancer Institute Surveillance, Epidemiology, and End Results Program. Accessed January 28, 2022. https://bit.ly/3s4vO9Y}

_{3. Llovet JM, Kelley RK, Villanueva A, et al. Hepatocellular carcinoma. Nat Rev Dis Primers. 2021;7(1):6. doi:10.1038/s41572-020-00240-3}

_{4. Kanwal F, Kramer J, Asch SM, Chayanupatkul M, Cao Y, El-Serag HB. Risk of hepatocellular cancer in HCV patients treated with direct-acting antiviral agents. Gastroenterology. 2017;153(4):996-1005.e1. doi:10.1053/j.gastro.2017.06.012}

_{5. Akinyemiju T, Abera S, Ahmed M, et al. The burden of primary liver cancer and underlying etiologies from 1990 to 2015 at the global, regional, and national level: results from the global burden of disease study 2015. JAMA Oncol. 2017;3(12):1683-1691. doi:10.1001/jamaoncol.2017.3055}

_{6. Anstee QM, Reeves HL, Kotsiliti E, Govaere O, Heikenwalder M. From NASH to HCC: current concepts and future challenges. Nat Rev Gastroenterol Hepatol. 2019;16(7):411-428. doi:10.1038/s41575-019-0145-7}

_{7. Estes C, Razavi H, Loomba R, Younossi Z, Sanyal AJ. Modeling the epidemic of nonalcoholic fatty liver disease demonstrates an exponential increase in burden of disease. Hepatology. 2018;67(1):123-133. doi:10.1002/hep.29466}

_{8. Sartoris R, Gregory J, Dioguardi Burgio M, Ronot M, Vilgrain V. HCC advances in diagnosis and prognosis: Digital and Imaging. Liver Int. 2021;41(S1):73-77. doi:10.1111/liv.14865}

_{9. Marrero JA, Kulik LM, Sirlin CB, et al. Diagnosis, staging, and management of hepatocellular carcinoma: 2018 practice guidance by the American Association for the Study of Liver Diseases. Hepatology. 2018;68(2):723-750. doi:10.1002/hep.29913}

_{10. Marks RM, Masch WR, Chernyak V. LI-RADS: past, present, and future, from the AJR special series on radiology reporting and data systems. AJR Am J Roentgenol. 2021;216(2):295-304. doi:10.2214/AJR.20.24272}

_{11. van der Pol CB, Lim CS, Sirlin CB, et al. Accuracy of the liver imaging reporting and data system in computed tomography and magnetic resonance image analysis of hepatocellular carcinoma or overall malignancy-a systematic review. Gastroenterology. 2019;156(4):976-986. doi:10.1053/j.gastro.2018.11.020}

_{12. Benson AB, D’Angelica MI, Abbott DE, et al. Hepatobiliary Cancers, Version 2.2021, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw. 2021;19(5):541-565. doi:10.6004/jnccn.2021.0022}

_{13. Abdel-Rahman O, Elsayed Z. Yttrium-90 microsphere radioembolisation for unresectable hepatocellular carcinoma. Cochrane Database Syst Rev. 2016;2:CD011313. doi:10.1002/14651858.CD011313.pub2}

_{14. Vouche M, Habib A, Ward T. Unresectable solitary hepatocellular carcinoma not amenable to radiofrequency ablation: multicenter radiology-pathology correlation and survival of radiation segmentectomy. Hepatology. 2014;60(1):192-201. doi:10.1002/hep.27057}

_{15. Yang B, Liang J, Qu Z, Yang F, Liao Z, Gou H. Transarterial strategies for the treatment of unresectable hepatocellular carcinoma: a systematic review. PLoS One. 2020;15(2):e0227475. doi:10.1371/journal.pone.0227475}

_{16. Llovet JM, Ricci S, Mazzaferro V, et al; SHARP Investigators Study Group. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med. 2008;359(4):378-390. doi:10.1056/NEJMoa0708857}

_{17. Bruix J, Qin S, Merle P, et al; RESORCE Investigators. Regorafenib for patients with hepatocellular carcinoma who progressed on sorafenib treatment (RESORCE): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet. 2017;389(10064):56-66. doi:10.1016/s0140-6736(16)32453-9}

_{18. Kudo M, Finn RS, Qin S, et al. Lenvatinib versus sorafenib in first-line treatment of patients with unresectable hepatocellular carcinoma: a randomised phase 3 non-inferiority trial. Lancet. 2018;391(10126):1163-1173. doi:10.1016/s0140-6736(18)30207-1}

_{19. Zhu AX, Finn RS, Edeline J, et al; KEYNOTE-224 investigators. Pembrolizumab in patients with advanced hepatocellular carcinoma previously treated with sorafenib (KEYNOTE-224): a non-randomised, open-label phase 2 trial. Lancet Oncol. 2018;19(7):940-952. doi:10.1016/s1470-2045(18)30351-6}

_{20. Finn RS, Ryoo BY, Merle P, et al; KEYNOTE-240 investigators. Pembrolizumab as second-line therapy in patients with advanced hepatocellular carcinoma in KEYNOTE-240: a randomized, double-blind, phase III trial. J Clin Oncol. 2020;38(3):193-202. doi:10.1200/jco.19.01307}

_{21. Meeting of the Oncologic Drugs Advisory Committee meeting announcement. FDA. April 29, 2021. Accessed January 28, 2022. https://bit.ly/2PyhdV1}

_{22. Qin S, Chen Z, Fang W, et al. Pembrolizumab plus best supportive care versus placebo plus best supportive care as second-line therapy in patients in Asia with advanced hepatocellular carcinoma (HCC): Phase 3 KEYNOTE-394 study. J Clin Oncol. 2022;40(suppl 4):383. doi:10.1200/JCO.2022.40.4_suppl.383}

_{23. Abou-Alfa GK, Meyer T, Cheng AL, et al. Cabozantinib in patients with advanced and progressing hepatocellular carcinoma. N Engl J Med. 2018;379(1):54-63. doi:10.1056/NEJMoa1717002}

_{24. NCCN Clinical Practice Guidelines in Oncology. Hepatobiliary Cancers. Version 5.2021. September 21, 2021. Accessed January 28, 2022. https://bit.ly/3INxv2m}

_{25. Zhu AX, Kang YK, Yen CJ, et al; REACH-2 study investigators. Ramucirumab after sorafenib in patients with advanced hepatocellular carcinoma and increased α-fetoprotein concentrations (REACH-2): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol. 2019;20(2):282-296. doi:10.1016/s1470-2045(18)30937-9}

_{26. El-Khoueiry AB, Sangro B, Yau T, et al. Nivolumab in patients with advanced hepatocellular carcinoma (CheckMate 040): an open-label, non-comparative, phase 1/2 dose escalation and expansion trial. Lancet. 2017;389(10088):2492-2502. doi:10.1016/S0140-6736(17)31046-2}

_{27. Yau T, Park JW, Finn RS, et al. Nivolumab versus sorafenib in advanced hepatocellular carcinoma (CheckMate 459): a randomised, multicentre, open-label, phase 3 trial. Lancet Oncol. 2022;23(1):77-90. doi:10.1016/S1470-2045(21)00604-5}

_{28. Bristol Myers Squibb Statement on Opdivo® (nivolumab) Monotherapy Post-Sorafenib Hepatocellular Carcinoma U.S. Indication. News release. Bristol Myers Squibb. July 23, 2021. Accessed January 28, 2022. https://bit.ly/3rXkSuR}

_{29. Yau T, Kang Y-K, Kim T-Y, et al. Efficacy and safety of nivolumab plus ipilimumab in patients with advanced hepatocellular carcinoma previously treated with sorafenib: the CheckMate 040 randomized clinical trial. JAMA Oncol. 2020;6(11):e204564. doi:10.1001/jamaoncol.2020.4564}

_{30. Finn RS, Qin S, Ikeda M, et al; IMbrave150 Investigators. Atezolizumab plus bevacizumab in unresectable hepatocellular carcinoma. N Engl J Med. 2020;382(20):1894-1905. doi:10.1056/NEJMoa1915745}

_{31. Finn RS, Qin S, Ikeda M, et al. IMbrave150: Updated overall survival (OS) data from a global, randomized, open-label phase III study of atezolizumab (atezo) + bevacizumab (bev) versus sorafenib (sor) in patients (pts) with unresectable hepatocellular carcinoma (HCC). J Clin Oncol. 2021;39(3_suppl):267-267. doi:10.1200/JCO.2021.39.3_suppl.267}

_{32. Chau I, Park JO, Ryoo BY, et al. Alpha-fetoprotein kinetics in patients with hepatocellular carcinoma receiving ramucirumab or placebo: an analysis of the phase 3 REACH study. Br J Cancer. 2018;119(1):19-26. doi:10.1038/s41416-018-0103-0}