What is proven, what is observed, what is not yet known

Arginine deprivation is not a new idea. It has fifteen years of laboratory work, dozens of clinical trials and one positive Phase III behind it. NGA-8000 is the newest enzyme built on that foundation. It has been given to more than 100 patients, and its documented record consists of five structured observations. This page separates what is proven from what is observed.

Fifteen years of arginine deprivation

2000sResearchers notice that melanoma and liver cancer cells die in culture when arginine is removed, while normal cells survive. The cause is traced to silencing of the gene ASS1.

2010A landmark review (Delage et al., International Journal of Cancer) maps ASS1 loss across tumour types and proposes arginine deprivation as a therapy.

2010 to 2020The bacterial enzyme ADI-PEG 20 enters trials in melanoma, liver cancer, mesothelioma, sarcoma, leukaemia and others. Tolerability is good; antibodies limit duration. A human arginase, pegzilarginase, is developed to overcome this.

2014 and 2022Laboratory work describes how starved cancer cells die: mitochondrial failure, DNA leakage, chromatin autophagy (Changou et al., PNAS 2014); apoptosis and cell-cycle arrest in gastric cancer under a thermostable arginase (2022).

2023Pegzilarginase is approved in the European Union for arginase-1 deficiency, a metabolic disease. Polaris submits ADI-PEG 20 for approval in mesothelioma.

2024ATOMIC-Meso, the Phase III trial of ADI-PEG 20 plus chemotherapy in pleural mesothelioma, reports a survival benefit (JAMA Oncology).

The principle in patients

The ATOMIC-Meso trial randomised 249 patients with non-epithelioid pleural mesothelioma, a cancer with a five-year survival of about 8 %, to chemotherapy with or without ADI-PEG 20. Median survival rose from 7.7 to 9.3 months, and the share of patients alive at three years roughly tripled. Side effects added by the enzyme were mostly injection-site reactions and mild blood count changes.

Earlier Phase II trials with ADI-PEG 20 showed disease control in liver cancer, melanoma and acute myeloid leukaemia, with the same pattern: responses in ASS1-negative tumours, loss of effect once antibodies appeared. Pegzilarginase has completed Phase I and II work in solid tumours, alone and with chemotherapy and anti-PD-1 therapy, with a manageable safety profile.

What this establishesDepriving a tumour of arginine is safe enough for months of treatment, and it extends life in at least one cancer when combined with chemotherapy. What it does not establish is the effect of NGA-8000 specifically.

How NGA-8000 differs

	ADI-PEG 20 (pegargiminase)	AEB1102 (pegzilarginase)	NGA-8000
Origin	recombinant bacterial protein (Mycoplasma)	recombinant human arginase 1, PEGylated	recombinant mammalian arginase, albumin-binding domain
Reaction	arginine → citrulline	arginine → ornithine	arginine → ornithine
Sensitive tumours	ASS1-negative	ASS1- or OTC-negative	ASS1- or OTC-negative
Immunogenicity	high; antibodies after ~8 weeks, loss of effect	low	low (our laboratory data, not independently verified)
Half-life	short	short to intermediate	extended by albumin binding (our laboratory data)
Clinical status	Phase III (ATOMIC-Meso, positive)	EU approval in ARG1 deficiency 2023; oncology Phase I/II	preclinical; individual cases; no approval
Combinations studied	cisplatin, pemetrexed, docetaxel, temozolomide, azacitidine, venetoclax	chemotherapy, anti-PD-1	cisplatin, gemcitabine, trastuzumab, anti-PD-1/PD-L1, radiotherapy (planned)

Compiled from our laboratory documentation and published literature. As of October 2026.

Early observations with NGA-8000

More than 100 patients have received NGA-8000 through the programme and its partner physicians. For five of them, laboratory values before and during treatment were recorded in a form we can show. No control group, partly alongside standard therapy, so effects cannot be attributed with certainty. No adverse events were reported in any of them.

Triple-negative breast cancer, stage IV

Woman, 50, metastatic. Seven weekly doses alongside chemotherapy, March to June 2025.

CEA (ref. < 5 ng/mL): elevated → within reference range by day 37CA 15-3 (ref. < 31.3 U/mL): within reference range by day 37

Chronic lymphocytic leukaemia, stage A

Man, 48. Five weekly doses, May to June 2025, no other treatment.

Lymphocytes: 51.1 → 40.0 × 10⁹/L in 27 daysChange −21.7 %; reference range not yet reached

Lung cancer, stage IV, bone metastases

2024.

CEA: 12 → 10.8 → 3.8 ng/mL over 4 months

Breast cancer recurrence, stage IV

2024.

CEA: 10.4 → 8.6 → 3.7 ng/mL in 1 month

Colorectal cancer, stage III

2024.

CEA decline documented; raw values with clinicians

Dose dates and raw data are available in the clinicians' section.

What we do not know

SurvivalNo study has measured whether NGA-8000 extends life. Falling tumour markers are a signal of activity, not a guarantee of outcome.

Response rateMore than 100 patients have been treated, but only five under uniform documentation. That is not enough for a response rate. We do not quote one, and we would distrust anyone who does.

Which tumoursASS1 and OTC loss predicts sensitivity in the laboratory. How well it predicts response in patients treated with NGA-8000 is unknown.

Long-term safetyRelated enzymes have been given for a year or more without serious harm. NGA-8000 itself has been observed over months, not years.

Our own dataHalf-life and immunogenicity data for NGA-8000 come from our own laboratory. They are consistent with the molecule's design, but they have not been verified by an independent group, and we say so.

Every patient in the programme is told this before consent, in these words.

Literature

Delage B. et al. Arginine deprivation and argininosuccinate synthetase expression in the treatment of cancer. Int J Cancer 2010;126:2762–2772.
Szlosarek P.W. et al. Pegargiminase plus first-line chemotherapy in patients with nonepithelioid pleural mesothelioma: the ATOMIC-Meso randomized clinical trial. JAMA Oncol 2024;10:475–483.
Changou C.A. et al. Arginine starvation-associated atypical cellular death involves mitochondrial dysfunction, nuclear DNA leakage, and chromatin autophagy. PNAS 2014;111:14147–14152.
Cheng P.N.-M. et al. Mono-PEGylated thermostable Bacillus caldovelox arginase mutant (BCA-M-PEG20) induces apoptosis, autophagy, cell cycle arrest and growth inhibition in gastric cancer cells. 2022.
Zhang Y. et al. Pegylated arginine deiminase drives arginine turnover and systemic autophagy to dictate energy metabolism. Cell Rep Med 2022;3:100498.
Tsai H.J. et al. A phase II study of arginine deiminase (ADI-PEG 20) in relapsed/refractory or poor-risk acute myeloid leukemia patients. Sci Rep 2017;7:11253.
Zou S. et al. Arginine metabolism and deprivation in cancer therapy. Biomed Pharmacother 2019;118:109210.
Dillon B.J. et al. Incidence and distribution of argininosuccinate synthetase deficiency in human cancers. Cancer 2004;100:826–833.
European Medicines Agency. Loargys (pegzilarginase), EPAR, 2023.

Laboratory documentation on NGA-8000 is available to clinicians on request.

See whether the evidence applies to your tumour type

Is this for me?