ALK-positive lung cancer is one of those weirdly specific corners of cancer biology that sounds obscure until you realize how much it changes treatment. In a small slice of lung cancers, the ALK gene gets fused to another gene, creating a permanently jammed-on growth signal. Instead of treating that with generic chemo and vibes, doctors can use ALK inhibitors that target the driver directly. This has been one of the cleaner success stories in precision oncology - not perfect, but real Schneider et al., 2023.
The problem is that tumors are rude. Give them one drug, and many eventually evolve resistance. Give them a second-generation ALK inhibitor like alectinib or brigatinib, and some tumors develop secondary ALK mutations, including the notorious G1202R mutation, which is basically the molecular version of changing the locks and pretending nobody's home Cooper et al., 2022; Gemelli et al., 2024.
That is why this paper matters. It is not trying to solve all of lung cancer. It is trying to answer a very practical question: what do you do when the cancer has already burned through stronger ALK drugs?
What DRAGON actually found
The DRAGON trial was single-arm and multicenter, conducted across 36 centers in China. Everyone got deulorlatinib. There was no direct comparison group, which means this is not a cage match against lorlatinib, chemo, or anything else. That limitation matters. Still, the readout was solid enough to get attention.
Among 158 evaluable patients, the objective response rate was 43.7%. Disease control was higher, and the median time before the cancer grew again was 11.1 months. The brain numbers are especially worth noticing because ALK-positive lung cancer has an annoying habit of spreading to the central nervous system. In patients with measurable CNS disease, more than half responded in the brain. For the small subgroup with G1202R, response looked even stronger: 62.5%, with disease control in all eight patients studied Huang et al., 2026.
That last number comes with a giant asterisk wearing a fake mustache. Eight patients is eight patients. Encouraging? Yes. Final verdict? Absolutely not.
The part hype headlines tend to mumble
The obvious comparison hanging over this study is lorlatinib, the current heavyweight in later-line ALK disease, especially when brain metastases are involved. Lorlatinib set a high bar for CNS activity, but it also earned a reputation for neurocognitive and mood-related adverse effects that can make treatment feel like a bad roommate moved into your frontal lobe NCI, 2024; Priantti et al., 2024.
Deulorlatinib's safety signal in DRAGON is one reason people will keep talking about it. Treatment-related adverse events were common, as usual, but dose reductions and discontinuations were relatively infrequent, and CNS-related toxicities did not appear to trigger interruptions or permanent discontinuation in the way clinicians often worry about with brain-penetrant ALK drugs Huang et al., 2026.
But let's not hand out a trophy made of premature certainty. This was not a randomized trial. It does not prove deulorlatinib is better than lorlatinib, safer than lorlatinib in a head-to-head sense, or the new universal answer after second-generation ALK inhibitor failure. It says something narrower, but still useful: here is a drug that looks active in a tough setting and did not immediately light the CNS toxicity alarm bell.
Why this could matter outside conference slides
If these results hold up, deulorlatinib could widen the menu for a group of patients who badly need options. That matters because real life is messier than treatment algorithms. Some patients progress in the brain. Some pick up resistance mutations. Some need something potent but more tolerable. Some have already had enough of being a one-person sequel franchise called Return of the Relapsing Tumor.
The broader ALK field is already moving toward smarter sequencing, better resistance testing with plasma DNA, and newer agents designed for stubborn mutations and brain disease Testa et al., 2024; Shen et al., 2023. Deulorlatinib fits that story. Not as magic. As pressure on a real bottleneck.
And honestly, that is often what progress looks like in cancer medicine. Not a trumpet blast. More like a locksmith showing up after the tumor changed the locks again.
References
Huang J, Chen G, Wang Z, et al. Deulorlatinib (TGRX-326) in ALK gene fusion positive non-small cell lung cancer after failure of second-generation inhibitors (DRAGON): a single-arm, multicenter, phase 2 trial. J Thorac Oncol. 2026. DOI: 10.1016/j.jtho.2026.103737
Schneider JL, Lin JJ, Shaw AT. ALK-positive lung cancer: a moving target. Nat Cancer. 2023;4(3):330-343. DOI: 10.1038/s43018-023-00515-0. PMCID: PMC10754274
Cooper AJ, Sequist LV, Lin JJ. Third-generation EGFR and ALK inhibitors: mechanisms of resistance and management. Nat Rev Clin Oncol. 2022;19(8):499-514. DOI: 10.1038/s41571-022-00639-9
Gemelli M, Albini A, Catalano G, et al. Navigating resistance to ALK inhibitors in the lorlatinib era: a comprehensive perspective on NSCLC. Expert Rev Anticancer Ther. 2024;24(6):347-361. DOI: 10.1080/14737140.2024.2344648
Shen CI, Lin CC, Ou SI. Treatment of advanced ALK-rearranged NSCLC following second-generation ALK-TKI failure. Expert Rev Anticancer Ther. 2023;23(11):1157-1167. DOI: 10.1080/14737140.2023.2265566
Testa U, Castelli G, Pelosi E. Alk-rearranged lung adenocarcinoma: From molecular genetics to therapeutic targeting. Tumori. 2024;110(2):88-95. DOI: 10.1177/03008916231202149. PMCID: PMC11005315
Priantti JN, Vilbert M, de Moraes FCA, et al. Neurocognitive adverse events related to lorlatinib in non-small cell lung cancer: a systematic review and meta-analysis. Cancers (Basel). 2024;16(14):2611. DOI: 10.3390/cancers16142611
NCI Cancer Currents Blog. Lorlatinib slows growth of ALK-positive lung cancers, may prevent brain metastases. July 10, 2024. Available at: https://www.cancer.gov/news-events/cancer-currents-blog/2024/lorlatinib-alk-positive-lung-cancer-initial-treatment
Disclaimer: The image accompanying this article is for illustrative purposes only and does not depict actual experimental results, data, or biological mechanisms.