News & Views

Immune-checkpoint inhibitors have revolutionized the management of hepatocellular carcinoma. Currently, anti-PD-(L)-1 antibodies combined with either bevacizumab or anti-CTLA4 antibodies are the standard of care for advanced-stage tumours. Now, two phase III studies (CheckMate 9DW and APOLLO) have reported positive survival results in the first-line setting, although with distinct implications for clinical practice.

The HARMONi-2 trial demonstrated a significant improvement in progression-free survival with ivonescimab, a bispecific PD-1×VEGF antibody, versus pembrolizumab in patients with PD-L1-positive advanced-stage non-small-cell lung cancer. Ivonescimab might enhance antitumour immunity by normalizing the tumour vasculature and reversing immunosuppression. Although promising, global applicability of ivonescimab awaits further validation, including overall survival results, in diverse populations.

Surrogate end points in drug regulation are thought to reduce the time required to bring new drugs to market; however, only a few of the drugs approved on the basis of these outcomes have subsequently demonstrated robust improvements in overall survival (OS). If the FDA and other regulators were to shift their priority to patient-centred outcomes, such as OS, I argue that such a shift would probably lead to fewer, but also a higher standard of drugs entering the market, potentially with faster approval decisions because novel therapies would initially be tested in later lines and in patients with a worse prognosis.

Advanced-stage hepatocellular carcinoma is currently treated with various anti-PD-(L)1 antibody-containing regimens. Now, a triplet combining the anti-TIGIT antibody tiragolumab with one of these regimens has demonstrated promising efficacy in a phase Ib/II trial, although these data will need to be confirmed. This study highlights the value of umbrella trials while also raising questions regarding the most effective immune-targeting strategies in patients with this disease.

In a recently reported study, a novel engineered oncolytic Newcastle disease virus encoding porcine α1,3-galactosyltransferase was evaluated in monkeys with CRISPR-induced primary hepatocellular carcinoma and in a phase I clinical trial. The virus induced hyperacute tumour rejection and an objective response rate of 35% in 20 evaluable patients. This approach highlights the promises and challenges of oncolytic virus drug development.

The thought-provoking SONIA trial showed that patients with hormone receptor-positive, HER2-negative advanced-stage breast cancer receiving deferred (second-line) versus first-line cyclin dependent-kinase (CDK) 4/6 inhibitors have noninferior progression-free survival after second-line treatment; such an approach also results in substantial cost savings. Herein, we discuss some important limitations of this trial and argue that, owing to their effect on overall survival, CDK4/6 inhibitors should remain the standard-of-care first-line therapy.

Recent data from the FELIX trial evaluating obecabtagene autoleucel in patients with relapsed and/or refractory B acute lymphoblastic leukaemia (R/R B-ALL) suggest that this novel intermediate-affinity CD19-directed chimeric antigen receptor (CAR) T cell therapy is associated with a reduced incidence of severe immune-mediated toxicities compared with other commercially available CAR T cell products. The increasing number of therapies available for B-ALL makes treatment selection and sequencing of therapies increasingly challenging.

DeepGEM, an artificial intelligence (AI)-based model, accurately predicts the presence of key genomic alterations in histological slides prepared from samples obtained from patients with lung cancer. This approach provides a cost-effective alternative to genomic testing, generates spatial mutation maps and might support personalized treatment strategies. Validated in diverse datasets, DeepGEM highlights the potential of AI to transform precision oncology and improve global healthcare equity.

Recent results from the phase III AQUILA trial demonstrate the benefit of fixed-duration monotherapy with daratumumab over observation in patients with high-risk smouldering multiple myeloma, which changes the early interception treatment landscape. Herein, we discuss how future studies could build on this success and pave the way to eradicating multiple myeloma before it starts.

As one of the first studies testing perioperative anti-PD-(L)1 antibodies in resectable non-small-cell lung cancer (NSCLC), NADIM now confirms, in its final report, impressive 5-year clinical outcomes and that a pCR following neoadjuvant therapy translates into improved long-term survival. These data support the development of novel, personalized treatments for locally advanced resectable NSCLC.

The DESTINY-Breast06 trial investigated earlier and broader use of trastuzumab deruxtecan in patients with metastatic hormone-receptor-positive breast cancer, and demonstrated improvements in progression-free survival over standard chemotherapy. These data provide a meaningful advance; however, this strategy might not be right for all patients, and careful consideration is recommended before blanket use.

A number of therapeutics that target mediators of signalling in the hypothalamic and brainstem regions that control appetite, ingestive behaviour, satiety, nausea and vomiting are starting to move the needle on cancer cachexia. However, clarification of meaningful clinical benefits for patients and the primary end points that should support regulatory approval of cachexia treatments is needed.

Adjuvant chemoradiotherapy has a proven survival benefit for patients undergoing upfront resection of gastric cancer compared with surgery alone. Now, data from the TOPGEAR trial demonstrate that adding radiotherapy to standard-of-care perioperative chemotherapy offers no additional survival benefit. I discuss the implications for treatment and future research.

The development of clinically relevant artificial intelligence (AI) models has traditionally required access to extensive labelled datasets, which inevitably centre AI advances around large centres and private corporations. Data availability has also dictated the development of AI applications: most studies focus on common cancer types, and leave rare diseases behind. However, this paradigm is changing with the advent of foundation models, which enable the training of more powerful and robust AI systems using much smaller datasets.

In the past 2 years, substantial improvements have been made in the management of advanced-stage EGFR-mutant non-small-cell lung cancer. Recent studies have suggested added benefit from the combination of third-generation tyrosine-kinase inhibitors with either chemotherapy or a bispecific antibody targeting EGFR and MET. Herein, we summarize these advances and their implications for clinical practice.

BCMA-directed chimeric antigen receptor T cell therapies and bispecific T cell engagers are moving to earlier lines of therapy in multiple myeloma. In addition, combination therapy with the BCMA-targeting antibody–drug conjugate belantamab mafodotin at first or subsequent relapse has the potential to improve survival of patients with this disease. This increasing number of therapeutic options makes treatment selection and sequencing increasingly complex.

The NETTER-2 trial provides the first phase III evidence that ¹⁷⁷Lu-dotatate is active as first-line therapy in patients with gastroenteropancreatic neuroendocrine tumours with a Ki-67 proliferative index of 10–55%. This approach is an important addition to the first-line therapeutic armamentarium, although treatment selection based on patient-specific and tumour-specific characteristics remains appropriate.

De-escalation of treatment for HER2⁺ breast cancer is a priority, given the increase in cure rates owing in part to improved HER2-targeted therapies. In this regard, the neoadjuvant approach provides the ideal platform to test less-intensive treatment regimens. Here we highlight a study that demonstrated the role of the metabolic response after dual HER2 blockade as a method of selecting patients who are most likely to benefit from chemotherapy-free neoadjuvant therapy.

The majority of patients with cancers of unknown primary have unfavourable outcomes when they receive empirical chemotherapy. The shift towards using precision medicine-based treatment strategies involves two options: tissue-agnostic or site-specific approaches. Here, we reflect on how cytology-based deep learning tools can be leveraged in these approaches.

The FDA has approved nanoliposomal irinotecan, 5-fluorouracil, leucovorin and oxaliplatin (NALIRIFOX) for patients with metastatic pancreatic adenocarcinoma on the basis of results from the NAPOLI 3 trial, in which this four-drug regimen improved overall survival relative to a doublet regimen. Here we discuss how, in the context of prior results from the PRODIGE 4 trial testing 5-fluorouracil, leucovorin, irinotecan and oxaliplatin (modified FOLFIRINOX), NALIRIFOX does not seem to raise the bar, but rather exposes patients and health-care systems to financial toxicities.