Provectus Provides Updated PV-10 Data on Metastatic Uveal Melanoma
Updated metastatic uveal melanoma data.
On Monday (November 13, 2023), Provectus provided updated data from the Company’s ongoing Phase 1 clinical trial of investigational cancer immunotherapy PV-10 (rose bengal sodium) for the treatment of uveal melanoma (UM) metastatic to the liver (mUM) (NCT00986661). mUM patients comprise about 45% of this basket study, which has treated hepatocellular carcinoma (HCC) and several different types of liver metastases (e.g., colorectal cancer [mCRC], pancreatic cancer [mPC/mPDAC], etc.).
Metastases from uveal melanoma appear in 8% to 32% of the patients during the first 5 years and in 50% of the patients at 10 years after diagnosis of the primary tumor [15,17,18]. When liver metastases develop, the prognosis is poor and life expectancy reduces to 6 to 11 months and only 15% of patients are alive after more than 1 year [19,20]. The liver is the first site of systemic metastasis for 89% to 95% of patients and the exclusive site of systemic metastasis in more than 50% of patients. In approximately half of the cases also the lungs (24%), bone (16%), and skin (11%) may be involved. Very rarely metastases are found in lymph nodes (10%) and brain (5%) [21-24]. { Elas et al. Towards Liver: Selected Aspects of Uveal Melanoma Metastasis. Clin Oncol. 2020; 5: 1738.}
Twenty-five mUM patients were treated in Provectus’s single-center study at MD Anderson Cancer Center (MDACC) in Houston, Texas, which has been led by Sapna Patel, MD. Up to three hepatic mUM tumors could be injected per PV-10 treatment cycle. Response assessments were performed at Day 28, and then every three months. Patients with additional, injectable, visceral hepatic mUM disease could have received additional cycles of PV-10 after Day 28. Eligible patients could have also received standard of care (SOC) immune checkpoint blockade (CB) with PV-10 treatment: PD-1 or CTLA-4 + PD-1. Enrollment and treatment for this mUM patient population are complete. Only survival follow-up continues.
PV-10’s clinical development program for mUM comprised:
Monotherapy PV-10,
Then PV-10 in combination with pembrolizumab (PD-1; Keytruda®), and
Then PV-10 in combination with ipilimumab (CTLA-4; Yervoy®) and nivolumab (PD-1; Opdivo®) (IN).
Sixteen patients received two or more PV-10 treatment cycles (64%). A median of two hepatic lesions were injected per patient (range one to 10). Twelve patients were CB-refractory (48%). Eleven patients received PV-10 + IN (44%). In the data update provided:
17 Stage M1a (68%) patients, 7 M1b (28%) patients, and one M1c (4%) patient were treated,
For all patients, median overall survival (mOS) was estimated to be 11.0 months,
For M1a patients, mOS was 30.6 months (2.6 years),
For M1a patients who received PV-10 + IN, mOS was 50.0 months (4.2 years),
Of nine patients assessed by PET-CT, four M1a patients achieved durable complete metabolic responses (CMRs),
These CMRs comprised one patient who received monotherapy PV-10 and three patients who received PV-10 + IN. CMRs received a median of 1.5 PV-10 treatment cycles (range 1-3), PV-10 injections to a median of 2 hepatic tumors (2-3), and, for three of them, a median of one dose of IN (1-5),
Two CMRs had hepatic-only disease. Two had extrahepatic disease: lung, bones, and soft tissue; lung and omentum,
All CMRs are alive after a median follow-up of 39 months (3.3 years), range 24.6-61.6 months (2.1-5.1 years), and mOS has not yet been reached, and
The CMRs represent 24% of M1a patients and 16% of all patients.
Other mUM data.
Prognostic treatment factors. We also reviewed radiology data, treatment history, PV-10 administration, and combination treatment to assess the potential impact of these characteristics on OS. We considered several potential prognostic factors by comparing PV-10 treatment “responders” (i.e., patients with an OS ≥ 12 months) to “non-responders” (i.e., an OS < 12 months):
Maximum lesion diameter (all sites),
Maximum hepatic lesion diameter,
Sum of lesion diameters (all sites),
Sum of hepatic lesion diameters,
Number of hepatic lesions ≥ 1.0 cm diameter,
Number of prior systemic or liver-directed treatment lines,
Elapsed time from primary diagnosis to metastatic diagnosis,
Elapsed time from metastatic diagnosis to commencement of PV-10,
Number of PV-10 treatment cycles,
Cumulative number of PV-10 injections, and
Number of hepatic lesions injected with PV-10.
Several potential factors differed significantly between responders and non-responders:
Maximum lesion diameter (all sites),
Maximum hepatic lesion diameter,
Sum of lesion diameters (all sites),
Sum of hepatic lesion diameters, and
Number of hepatic lesions ≥ 1.0 cm diameter.
Among other things, we found that:
Maximum lesion diameter (all lesions) (MLD) yielded a significant OS difference for American Joint Committee on Cancer (AJCC) staging criteria: M1a versus M1b/M1c,
Sum of lesion diameters (hepatic lesions) (SLD) yielded a highly significant OS difference for a 10.0 cm threshold, and
Number of measurable hepatic lesions (≥ 1.0 cm) (N) yielded a highly significant OS difference for a 5-lesion threshold.
We also conducted an analysis combining certain potential prognostic factors, such as MLD-all lesions/SLD-liver/N, MLD-liver/SLD-liver/N, and SLD-liver/N.
Disease burden matters to prognosis following PV-10 treatment and relatively simple criteria can be used to select patients with the best prognosis for PV-10 treatment. AJCC staging (which is standard assessment criteria) of M1a patients versus M1b/M1c patients displayed high statistical significance and the greatest differential in median OS. The combination of PV-10 + IN for M1a (and M1b) patients trended favorably for OS and applied to both CB-naïve and CB-refractory patients.
DAMPs. Damage-associated molecular pattern molecules (DAMPs) and related cytokines can have immune up- and down-regulatory roles in cancer. Their pattern and levels in peripheral blood may be indicative of disease burden, immune competence, and therapeutic effect. We have shown that PV-10 treatment leads to fingerprint-like innate signaling and functional immune response. This DAMP-directed immune correlative assessment of PV-10-treated mUM patients is a key part of PV-10’s regulatory story.
Peripheral blood from mUM patients was analyzed for several DAMPs: HMGB1, HSP90a, S100A8, S100A9, S100A12, S100B, and IL-1a. Samples were generally collected at baseline (i.e., immediately prior to initial PV-10 administration) and/or at one or more follow-up intervals. We analyzed the DAMPs data for relationships, including baseline DAMPs versus OS or disease burden, change in DAMPs versus OS or PV-10 dose, and the chronology of the change in DAMPs. We observed that:
Baseline DAMPs appeared to be prognostic for OS,
DAMP release after PV-10 treatment appeared to occur in a predictable manner and was consistent with preclinical models and clinical data from other cancer indications involving PV-10 treatment, and
Increased DAMP release in high-dose patients implied a dose response.
We also noted that while collecting blood samples for DAMP analysis at Day 28 is important, increased sensitivity should come from earlier collection, such as one to 14 days post-PV-10 treatment. Nonetheless, the data support PV-10’s immuno-oncology cycle and its potential influence on OS.
Timeline.
Provectus’s mUM program has worked through several issues and taken time to coalesce, from mid-2018 when the program began enrollment to late-2023, wrapping up this part of its clinical work. Among other aspects, we worked through the PV-10 dosing regimen, treatment response to monotherapy PV-10, treatment response to PV-10-led combination therapy, durability of such responses, survival, and mechanism of immune action.
Ipilimumab + Nivolumab. During this time, key events included the publication of data in February 2021 (Pelster et al.) of IN for mUM patients in an MDACC single-center, single-arm, Phase 2 study. Of 33 evaluable patients (35 total patients):
17 Stage M1a (49%) patients, 14 M1b (40%) patients, and four M1c (11%) patients were enrolled,
Objective response rate (ORR) was 18%, consisting of one complete response (CR) and five partial responses (PRs),
Median progression-free survival (mPFS) was 5.5 months,
mOS was 19.1 months,
40% of patients experienced a grade 3-4 treatment-related adverse event, and
20% had extrahepatic-only disease.
We analyzed the Pelster et al. data, comparing mOS of PV-10 + IN and IN for M1a patients with hepatic disease-only and hepatic/extrahepatic disease: 50.0 months versus 12.8 months.
Tebentafusp. Immunocore’s tebentafusp (Kimmtrak®) was approved in January 2022 for HLA-A*02:01-positive adult patients with unresectable or metastatic UM:
mOS was 21.7 months for patients treated with tebentafusp and 16 months in the investigator’s choice arm/control group of monotherapy pembrolizumab, ipilimumab, or chemotherapy dacarbazine (DTIC) (see Nathan et al. published in September 2021),
One-year OS was 73% for tebentafusp and 59% for the control,
ORR was 9% for tebentafusp and 5% for the control. Disease control rate (DCR) was 46% for tebentafusp and 27% for the control,
Three-year follow-up, published in October 2023 (Hassel et al.) noted that mOS was 21.6 months (lower, compared to the approval data) for tebentafusp and 16.9 months for the control (higher). The estimated percentage of patients surviving at three years was 27% for tebentafusp and 18% for the control, and
4% had extrahepatic-only disease.
We plan to analyze the Nathan et al. data to compare mOS for PV-10 + IN and tebentafusp in patients with hepatic-only and hepatic/extrahepatic disease. Accessing these data has been challenging; however, it would appear from the NEJM articles that tebentasfusp’s mOS for the target patient population is unlikely to exceed PV-10 + IN’s mOS.
Notably, the 2021 NEJM publication that supported tebentafusp’s approval was predicated on one-year OS that had a hazard ratio (HR) of 0.51 (73% versus 59%). The 2023 NEJM paper reported an HR of 0.68, mOS of 21.6 versus 16.9 months and three-year OS of 27% versus 18%. Tebentafusp’s effect size shrank markedly after longer follow-up.
To put this in context, ipilimumab’s cutaneous melanoma approval was based on an HR of 0.72 against DTIC (11.2 versus 9.1 months) (see Robert et al. NEJM 2019). Pembrolizumab was approved for ipilimumab-naïve cutaneous melanoma based on an HR of 0.63-0.69 versus ipilimumab (13.4-14.7 versus 11.0 months) (KN-006). Tebentafusp’s initial HR of 0.51 for one-year OS was remarkable; however, 0.68 for three-year OS was incremental.
Longer-term survival data for monotherapy and combination therapy PV-10 treatment of mUM, HCC, mCRC, mPC, cutaneous melanoma, metastatic neuroendocrine tumors, and other indications have been very informative and important to developing the next part of PV-10’s clinical program directed towards approval pathways.
RP2 + Nivolumab. Replimune presented data in November 2023 of its engineered oncolytic virus RP2, which is encoded with a human anti–CTLA-4 antibody-like molecule, as a monotherapy and in combination with nivolumab (i.e., akin to IN) at the Society of Melanoma Research annual meeting. For RP2 + nivolumab combination therapy:
29% ORR and 64% DCR,
No determination of mOS. For all 17 patients, mOS appeared to be approximately 460 days (15.1 months); however, for patients receiving RP2 + nivolumab (14 patients; three received monotherapy RP2), mOS appears to be immature,
11% of all patients and 14% of combination therapy patients had extrahepatic-only disease, and
Two patients with the longest survival were combination therapy patients who had extrahepatic-only disease.
We believe that PV-10-led treatment can beat standard of and emerging care for mUM.
Promise and proof.
The promise of PV-10 treatment was, and remains, the high amount of tumor response to PV-10 injection (i.e., complete response). The premise is simple, “touch and treat:”
Touch: Initial PV-10 injection of available disease for injection kills disease that…
Then leads to cell death (consequentially immunogenic cell death, but also other types of cell death; see Galluzi et al., “Molecular mechanisms of cell death: recommendations of the Nomenclature Committee on Cell Death 2018”) that…
Then leads to unique, multivariate innate signaling that…
And treat: Then leads to disease-specific T cell response.
The eventual fulfillment of the promise would be significant, durable, overall patient-level response (i.e., complete response) leading to long-term survival. Provectus began to advance this promise in its Phase 2 trial of monotherapy PV-10 for the treatment of advanced cutaneous melanoma, where a sub-group of patients ostensibly had all of their disease treated (i.e., injected with PV-10): “Efficacy of intralesional rose bengal in patients receiving injection in all existing melanoma in phase II study PV-10-MM-02,” 2014 American Society of Clinical Oncology. The sub-group achieved 50% CR and 71% ORR; however, data deficiencies and gaps included:
Modified RECIST (mRECIST) for response measurement,
Unclear differentiation between, and thus communication and presentation of, lesion-level and overall patient-level RECIST response,
Durability of response that was not measured long enough,
OS that was not measured long enough, hampering the potential connection of high lesion-level response to high patient-level response to survival, and
An undeveloped immune story to communicate the mechanism of immune action that would have connected local treatment to systemic benefit.
One could have hypothesized that PV-10 treatment-derived patient CR was prognostic of OS, and thus a better surrogate than PFS for OS; however, one would not have had all of the data to robustly and holistically prove it.
The use of PET-CT and PERCIST in a RECIST world.
The mUM CMRs were eye opening for the magnitude of patient outcome (i.e., high response) and the connection to long-term survival (i.e., high prognostic value). Comparably eye opening was the potential import of prognostic treatment factors (i.e., M1a staging) and further substantiating PV-10’s systemic immune story (i.e., DAMPs).
A positron emission tomography (PET)-computed tomography (CT) (PET-CT) scan combines the two types of scans (i.e., a PET scan and a CT scan) to provide detailed information about the structure and function of organs and tissues in the body. The combination of information from both PET and CT scans allows doctors to obtain a more comprehensive understanding of a patient’s condition. A PET-CT scan can help, among other things, to detect and evaluate cancer, determine the extent of disease, assess treatment response, and potentially plan further or alternative treatment strategies. The use of PET-CT scans is longstanding, informative, detailed, and potentially expensive.
Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 is the most used criteria to assess tumor response; however, RECIST measurement, which is based on CT, has limitations. PET Response Criteria in Solid Tumors (PERCIST) is an alternate means of assessing tumor response. PERCIST may provide more accurate information of therapeutic response at an earlier stage of treatment. RECIST- and PERCIST-based tumor responses, however, can be considerably different, with the former underestimating response measured by the latter.
Attempting to use PERCIST, alone and without context, to convince regulators, industry, and investors about PV-10’s promise is not too different from historically using mRECIST. PV-10 lives in a RECIST world, and RECIST data must accompany PERCIST data, irrespective of concordance and/or discordance, to comply with customary methodologies.
While Provectus’s mUM data suggest the potential prognostic power of CMR for OS, we must frame the data argument in unequivocal terms:
Lead with OS,
Present CMR (PERCIST), and
Confirm with ORR (RECIST) and, to a lesser extent, PFS (RECIST).
Our clinical trial protocols of Provectus’s part two of its clinical development program for PV-10 in mPDAC and mUM provide for PERCIST measurement starting six months after the start of PV-10 treatment.
Lessons learned about mUM and PV-10.
1. Enough PV-10 matters to patients.
More than one cycle of PV-10 may be needed. More than one lesion may need to be injected (and/or re-injected) with PV-10.
2. PV-10-injectable hepatic disease matters to patients.
The liver is the principal site of UM metastasis, substantially higher than metastatic spread to the lungs, bone, skin, or other sites. mUM patients with hepatic disease (and patients with liver metastases in general) have poorer prognoses that those with extrahepatic-only disease. We discussed this strategic approach to patient treatment in the July 24, 2023 Substack post titled “Statistical modeling of PV-10 and chemotherapy for FOLFIRINOX-refractory pancreatic cancer metastatic to the liver.”
3. Disease stage matters to patients.
Treating an earlier/the earliest stage of metastatic spread (i.e., M1a, but still Stage IV), where tumor burden is relatively lower, means there is greater opportunity for PV-10-led treatment to be a difference maker, both in terms of liver-directed treatment itself and the robustness of the systemic immune response.
4. Local PV-10 treatment mounts a systemic immune effect.
The late-2000s and early-2010s criticism of the key deficiency of localized, intratumoral PV-10 treatment of not being capable of mounting a systemic benefit for the patient are largely behind the Company. Behind Provectus is also the mid-2010s praise of PV-10 as the perfect way to prime immune system (see Weber 2014); PV-10 is much more. The work of the Company and its medical collaborators of the last several years has demonstrated multivariate innate signaling (e.g., DAMPs, STING, etc.) that connects “front-end” multifaceted cell death to “back-end” tumor specific T cell response.
5. PV-10 can lead to CMR, which in turn may lead to long-term survival.
PV-10-led combination therapy (and monotherapy PV-10) can help patients achieve CMR, which may mean a potential durable remission from cancer. Can PV-10 lead to five or more years of no evidence of disease? CMRs may potentially predict long-term survival. Is CMR a better surrogate for OS than PFS?
Clinical development program (CDP) notes.
Metastatic pancreatic ductal adenocarcinoma (mPDAC) program. mPDAC should lead regulatory-wise (i.e., with FDA) and clinical strategy-wise (i.e., “lead indication); see June 21, 2023 Substack post titled “Provectus’s 2023 Annual Stockholder Meeting” [2023 ASM]). Further and illustratively, the second-line gem + nab mPDAC one-year mOS rate of ~23% (see July 24, 2023 Substack post titled “Statistical modeling of PV-10 and chemotherapy for FOLFIRINOX-refractory pancreatic cancer metastatic to the liver”) is substantially lower than the mUM one-year mOS rates of tebentafusp (73%), its control group (50%), or single-arm IN (56%).
mUM. The mUM program would draft behind the mPDAC program regulatory- and clinical strategy-wise. The follow-on dedicated single-center study cohort would be M1a patients receiving PV-10 + IN, measure patient response by PERCIST (to seek a “high batting average;” as high a proportion of CMRs as possible) and key endpoints by RECIST, potentially further correlate CMR with OS, and further establish PV-10’s immunotherapy story.
Advanced cutaneous melanoma program. The Substack post to follow today’s post would provide updated data on PV-10 for this indication (press released on Thursday, November 16, 2023) and commentary, such as the relative priority of this program compared to mPDAC and mUM (an adjustment from the 2023 ASM presentation).
CDP, part two. As we discussed at the Company’s 2023 ASM, Provectus needs to ensure suitable funding is in place to effectively execute Part Two of the Company’s CDP for PV-10.
CDP, part one. Work to date transitioning Part One of the CDP has focused more on collecting survival data than just treating patients for response. Survival from PV-10 treatment has been very important to understand and assess, especially given Provectus’s historical PV-10 under-dosing (i.e., under-treating of patients).
A Phase 2 by any other name. There were and are pros, cons, and other considerations to undertaking the original and follow-on mUM patient study as either (i) a Phase 2 clinical trial (given the Phase 1 basket study’s precedent safety, tolerability, and dosing data) or (ii) a continuation of the Phase 1 trial with a dedicated mUM expansion cohort for when Provectus was ready to advance development given a meaningful therapeutic signal. All in all, it was ostensibly cheaper and more efficient (i.e., from a site activation, operations, cost, and other perspective) to do and continue the mUM program as a Phase 1 trial; however, we do understand questions and debate over study phases, phase numbers, etc.
Forward-Looking Statements
The information in Provectus’ Substack post may include “forward-looking statements,” within the meaning of U.S. securities legislation, relating to the business of Provectus and its affiliates, which are based on the opinions and estimates of Company management and are subject to a variety of risks and uncertainties and other factors that could cause actual events or results to differ materially from those projected in the forward-looking statements. Forward-looking statements are often, but not always, identified by the use of words such as “seek,” “anticipate,” “budget,” “plan,” “continue,” “estimate,” “expect,” “forecast,” “may,” “will,” “project,” “predict,” “potential,” “targeting,” “intend,” “could,” “might,” “should,” “believe,” and similar words suggesting future outcomes or statements regarding an outlook.
The safety and efficacy of the agents and/or uses under investigation have not been established. There is no guarantee that the agents will receive health authority approval or become commercially available in any country for the uses being investigated or that such agents as products will achieve any particular revenue levels.
Due to the risks, uncertainties, and assumptions inherent in forward-looking statements, readers should not place undue reliance on these forward-looking statements. The forward-looking statements contained in Provectus’ Substack are made as of the date hereof or as of the date specifically specified herein, and Provectus undertakes no obligation to update or revise any forward-looking statements, whether as a result of new information, future events or otherwise, except in accordance with applicable securities laws. The forward-looking statements are expressly qualified by this cautionary statement.
Risks, uncertainties, and assumptions include those discussed in the Company’s filings with the Securities and Exchange Commission (SEC), including those described in Item 1A of:
the Company’s Annual Report on Form 10-K for the year ended December 31, 2022, and
Provectus’ Quarterly Report on Form 10-Q for the period ended September 30, 2023.