JX-594 in Melanoma: Five Evaluable Patients, Big Questions

원문 링크: WordPress 원문

TRIAL REGISTRY · EN CLINICALTRIALS.GOV LOCK

Ten enrolled. Five evaluable. No CR or PR.

Bottom line: the newly posted ClinicalTrials.gov results do not support the strong response narrative previously attached to this trial. Ten patients entered the single-arm Phase I/II study, but only five were evaluable for efficacy. In those five, the registry reports no complete or partial responses across the entire disease burden and no complete or partial responses in non-injected tumors. Median progression-free survival could not be calculated. The study remains informative as an early safety and feasibility record, but it cannot establish systemic anti-tumor efficacy.

10

participants enrolled

5

efficacy-evaluable

CR: 0 · PR: 0

whole-disease response

ENG · English / KOR · 한국어 BILINGUAL REGISTRY BRIEF

DESIGN

Open-label single-group Phase I/II

REGIMEN

Intratumoral JX-594 weekly ×6

FIRST ASSESSMENT

Day 43

01 If You Remember Only One Thing

Of 10 enrolled patients with unresectable Stage 3/4 malignant melanoma, 5 were evaluable for the efficacy analysis. Among those 5 patients, no one achieved a complete or partial response — for whole disease burden, 3 had stable disease and 2 had progressive disease. Stable disease is not an objective response. The registry’s primary outcome field carries an internal labelling mismatch that is disclosed in this article. This record does not support a systemic anti-tumour efficacy claim for JX-594.

02 Terms in Plain Language

Viral vector: A virus modified in a laboratory to carry a biological payload or instruction. In this study, the vector was a thymidine-kinase-inactivated vaccinia virus engineered to express human GM-CSF. The registry record identifies that design; it does not by itself prove how selectively the vector behaved in each patient’s tissues.

JX-594: A modified vaccinia virus, also called Pexa-Vec. Its thymidine kinase activity was inactivated, and it was engineered to express human GM-CSF. These are design features reported in the registry. Whether the intended biological steps occurred in these patients cannot be established from the posted outcome tables alone.

GM-CSF (Granulocyte-Macrophage Colony-Stimulating Factor): An immune signalling protein. JX-594 was engineered to express human GM-CSF, forming part of the treatment’s biological rationale. The registry result tables do not directly measure or prove local GM-CSF expression or immune recruitment in these participants.

Intratumoral injection: The drug was injected directly into the tumour using a needle, rather than being swallowed as a pill or delivered into a vein. Intratumoral delivery concentrates the agent at the target site but does not guarantee that effects will reach other tumour sites in the body.

Injected tumour: The specific tumour lesion that received the direct injection of JX-594. Only this lesion received the drug directly.

Non-injected tumours: Other tumour lesions elsewhere in the body that did not receive any direct injection. Observing the behaviour of non-injected tumours is one way researchers look for signals of a systemic (whole-body) effect from a locally delivered treatment.

Whole disease burden: All tumour lesions in the body considered together — injected and non-injected combined — as a single overall assessment.

CR, PR, SD, PD: Standard oncology response categories. Complete response (CR) means all detectable cancer disappeared. Partial response (PR) means the tumour shrank by a defined threshold. Stable disease (SD) means the tumour neither grew significantly nor shrank significantly over the observation window. Progressive disease (PD) means the cancer grew or new lesions appeared. CR and PR together constitute an objective response. SD is not an objective response.

PFS (Progression-Free Survival): The time from enrolment until cancer progression or death. In this registry record, the median PFS and its confidence limits are posted as not available (NA) and cannot be calculated from the available data.

Denominator: The count of patients whose data are included in a specific calculation. The denominator for the efficacy results in this trial is 5 — not the 10 who were enrolled — because only 5 patients met the criteria for the evaluable population.

03 Why the Study Was Attempted

The trial enrolled adults with unresectable Stage 3 or Stage 4 malignant melanoma and at least one measurable tumour that could be injected directly or with ultrasound guidance. The central practical idea was to place the investigational biological agent inside an accessible lesion and then observe local, distant, and whole-disease outcomes separately.

JX-594 was designed as a thymidine-kinase-inactivated vaccinia virus expressing human GM-CSF. In simple terms, the treatment concept combined direct delivery of a modified virus into a tumour with an immune-signalling payload. That explains why the protocol tracked injected tumours and non-injected tumours as different outcome scopes.

The possibility that a locally injected treatment could influence distant, untreated lesions was a hypothesis to test, not a result to assume. This small, non-comparative early-phase study could look for a preliminary signal, but it could not by itself confirm systemic efficacy.

04 Who Was Treated and Study Design

Ten patients with unresectable Stage 3/4 malignant melanoma were enrolled. JX-594 was administered intratumorally at a dose of 1 × 10 8 plaque-forming units (pfu) once weekly for 6 treatments delivered over 6 weeks. The trial was open-label — patients and investigators knew the treatment being given — and used a non-comparative single-group design. There was no control arm.

The trial started enrolment in February 2007. Primary completion was February 2008. Full study completion was December 2009. Registry results were first posted on 2026-07-27. The trial identifier is NCT00429312.

05 Data at a Glance

Enrolment and completion (completion/dropout module):

• Enrolled: 10

• Completed: 5

• Did not complete: 5

• Reasons for non-completion: withdrawal by subject (1), physician decision (1), incomplete efficacy evaluation (3)

PFS population exclusion reasons (separate module, separately scoped):

• No CT scan: 1

• Efficacy assessments more than 28 days beyond the protocol window: 2

• Early withdrawal before Day 43: 2

These two descriptions come from differently scoped registry modules and reflect different grouping criteria. They are not a single reconciled list and should not be merged into one.

Efficacy results (denominator = 5 evaluable patients):

• Whole disease burden: CR 0, PR 0, SD 3, PD 2

• Non-injected tumours: CR 0, PR 0, SD 3, PD 2

• PFS: Denominator 5; median and confidence limits posted as NA — not calculable

Primary outcome registry note — disclosed mismatch: The primary outcome field title reads “Number of Participants With Objective Response in Injected Tumor(s).” The posted result unit, however, is “Participants with Stable disease,” with a value of 5 of a denominator of 5. Stable disease is not an objective response. This is an internal labelling inconsistency within the registry record. It is disclosed here rather than resolved by substituting one label for the other.

AI-generated conceptual explainer and not study data. Diagram illustrating the JX-594 oncolytic virus approaching a tumour cell, along with local expression of GM-CSF recruiting immune cells. Key labels depicted: JX-594, Tumor cell, GM-CSF.

06 How to Read Each Result

Denominator is 5, not 10: Ten patients were enrolled, but only 5 met the criteria for the evaluable efficacy population under the PFS population definition. The efficacy results — whole disease burden, non-injected tumour response, and PFS — all use 5 as the denominator. Results should not be reported as fractions of 10 in the efficacy context.

Stable disease (SD) is not an objective response: Three of 5 evaluable patients had stable disease for whole disease burden. SD means the cancer did not grow or shrink significantly during the observation window. This does not represent tumour shrinkage or elimination and cannot be described as an objective response or as a therapeutic success signal in the way that CR or PR can.

Progressive disease (PD): Two of 5 evaluable patients had progressive disease — their cancer grew or new lesions appeared during the observation period.

CR and PR: Zero patients in the evaluable population achieved either a complete or partial response for whole disease burden or for non-injected tumours.

PFS: The registry posted the median and confidence interval limits as NA. No PFS figure can be derived from this record.

Primary outcome mismatch — how to read it: The registry field labelled as the primary outcome says “Objective Response in Injected Tumor(s),” but the data value entered beneath it is “Participants with Stable disease” at 5 of 5. Reading the data value (stable disease) as if it corresponded to the field title (objective response) would misrepresent what was observed. Both the title and the result are reported here exactly as they appear in the registry, with the labelling discrepancy disclosed.

AI-generated conceptual explainer and not study data. Diagram illustrating the distinction between an injected tumour (receiving direct JX-594 injection), non-injected tumours (distant lesions not directly treated), and whole disease burden (all lesions combined). Key labels depicted: Injected tumor, Non-injected tumor, Whole disease burden.

07 Mechanism and What the Data Do Not Prove

JX-594 combined direct intratumoral delivery of a modified vaccinia virus with a design intended to express human GM-CSF. One important study question was therefore whether changes would appear not only in the directly injected lesion but also in non-injected lesions and whole disease burden. That was a hypothesis to test, not a result established by the vector design.

The data in this registry record do not demonstrate that this mechanism operated in these patients. No complete or partial response was recorded in non-injected tumours or in whole disease burden. Three of 5 evaluable patients had stable disease at non-injected sites. Stable disease alone does not establish a treatment-attributable effect at a distance.

Whether JX-594 infected or lysed any tumour cells in these patients, whether GM-CSF was locally expressed, and whether any immune recruitment occurred are mechanistic questions that cannot be answered from registry outcome data alone.

08 Safety with Attribution

The registry safety ledger records treatment-related adverse events (AEs) in 7 of 10 patients (through Day 64). Treatment-related serious adverse events (SAEs) were recorded in 0 of 10 patients through the same period.

The registry’s all-cause adverse event module records separately: all-cause serious adverse events in 2 of 10 patients, and all-cause other adverse events in 10 of 10 patients. The two all-cause serious events were acute gastrointestinal bleeding and spinal cord compression.

These two events — acute gastrointestinal bleeding and spinal cord compression — appear in the all-cause serious event module, not in the treatment-related SAE module. The treatment-related SAE denominator is 0 of 10. These events must not be described as treatment-related based on the posted registry data.

09 What Can and Cannot Be Concluded

Supported by this registry record:

• Among 5 evaluable patients, 0 achieved CR or PR for whole disease burden or for non-injected tumours.

• 3 of 5 evaluable patients had stable disease; 2 of 5 had progressive disease (whole disease burden).

• Treatment-related AEs were recorded in 7 of 10 patients; treatment-related SAEs in 0 of 10 (through Day 64).

• The primary outcome field contains an internal labelling mismatch between its title and its posted result unit.

• PFS median and confidence limits are posted as NA.

Not supported by this registry record:

• That JX-594 produced an unambiguous objective-response result in injected tumours — the primary field’s title and posted unit conflict, so the injected-tumour result cannot safely be converted into an objective-response claim.

• That JX-594 produced objective responses in non-injected tumours — no CR or PR is recorded.

• That systemic anti-tumour activity occurred — the denominator-5 non-injected tumour results show no objective responses.

• That the two all-cause serious AEs (GI bleeding, spinal cord compression) were treatment-related — the registry records 0 treatment-related SAEs.

• That a PFS value can be calculated — posted as NA.

• That stable disease represents an objective response — it does not.

10 Evidence Boundary

This article is based on a ClinicalTrials.gov registry record (NCT00429312) only. Registry records are structured data submissions and may contain labelling inconsistencies, as illustrated by the primary outcome mismatch described above. This record is not a peer-reviewed publication; it does not contain patient-level details, statistical methodology descriptions, or narrative clinical discussion. The trial enrolled 10 patients in 2007–2008 in a single-group early-phase design; its results cannot be extrapolated to the broader melanoma population or to later generations of oncolytic virus approaches. Registry results were first posted on 2026-07-27. No peer-reviewed publication of these specific results was referenced in this article.

11 Official References

• ClinicalTrials.gov study record: NCT00429312

• ClinicalTrials.gov API record: NCT00429312

12 Medical Information Only — Not Medical Advice

This article presents publicly available registry information for educational purposes only. It is not a substitute for professional medical advice, diagnosis, or treatment recommendations. Decisions about cancer care must be made in consultation with qualified healthcare providers who know each patient’s full clinical history and circumstances. The registry record described here is from an early-phase study conducted in 2007–2008; its data should not be interpreted as guidance for any individual treatment decision. The primary outcome labelling mismatch described in this article has been disclosed as-found; the underlying clinical significance of the data requires interpretation by qualified professionals.