May 2020 Edition Vol.11, Issue 5

Live (from home), it’s AACR Virtual I!

By Neil Canavan

The annual meeting of the American Association for Cancer Research (AACR) hosts, on average, some 20,000 clinicians, scientists, and related stakeholders all merrily packed into plenaries sessions, poster halls, and press rooms. This year, a record-breaking 60,000 individuals registered for the rescheduled, abbreviated, two-day AACR Virtual Annual Meeting I. And no one came. Instead, one of the largest medical meetings of its kind went online, and what was gained, and some of what was lost, is explicated herein. 

Immunotherapy Triplets: Breast, Melanoma

Since their much heralded advent, the class of drugs known as checkpoint inhibitors have been combined with almost everything, and it almost never works. This time it did. The I-SPY 2 trial combined the IO/anti-PD-L1 drug, durvalumab, with the PARP inhibitor, olaparib, and the chemotherapeutic paclitaxel. 

The IO portion had already shown activity when combined with both a taxane and an anthracycline in the neoadjuvant setting, where a pathologic complete response (pCR) of 44% was observed. Whereas the PARP inhibitor, olaparib, had previously demonstrated a response rate (RR) of 60% in the difficult to treat, BRCA-positive, triple negative breast cancer (TNBC) population.

“We expected synergy between these two classes of agents,” said study presenter, Lajos Pusztai, MD, DPhil, of Yale University, with the PARP inhibitor contributing to this synergy in two ways. First, olaparib inhibits a DNAdamage repair enzyme, which induces a greater variety of genetic mutations for the immune system to recognize and attack; secondly, by upregulating the expression of PD-L1 in the tumor microenvironment, olaparib enables the immune-boosting activity of durvalumab. 

For the phase II, I-SPY 2 study, investigators enrolled HER2-negative breast cancer patients who were either HR-positive, or HR-negative (TNBC). Patients were then randomized to 12 weeks of neoadjuvant treatment with either the triplet (N=73) or the control arm of paclitaxel (N=299) with all patients then undergoing 8-12 weeks of doxorubicin/cyclophosphamide, followed by surgery. The primary endpoint of I-SPY 2 was achievement of pCR.

The results, in brief: “There were no unexpected safety signals,” Pusztai reported, although there was more febrile neutropenia in the experimental arm (16.3% vs 8.4%) as well as 19% of patients experiencing some immune-related events due to the activity of durvalumab. 

As for efficacy, the triplet combination increased pCR rates in all three biomarker subsets. For HER2-negative patients pCR was 37% vs 22% for the control, in HER2-negative, ER positive, mamma print high risk patients, it was 28% vs 14%, and for TNBC patients it was a 47% vs 27% for the control group. “The estimated probability that the experimental arm is superior to chemotherapy alone is greater than 98% in all subsets,” Pusztai concluded.

Melanoma: IMspire150

Grant McArthur, PhD, of the MacCallum Cancer Center, Melbourne, presented the preliminary results from the phase III, IMSpire150 trial that looked at the combination of the IO drug atezolizumab (AZO), the MEK inhibitor, combimetinib (COM), and the BRAFv600 mutant inhibitor, vemurafenib (VEM) in treatment naïve, BRAFV600-positive advanced melanoma patients.

“Patients with BRAF melanoma can be treated with combinations of BRAF and MEK inhibitors, and offer the advantage of high response rates,” said McArthur. “But unfortunately, in many patients these responses are short lived. On the other hand, immune checkpoints inhibitors have more durable responses, but have a lower response rate.” These observations, plus the hypothesis that BRAF and MEK inhibitors seem to drive T cells into tumors as well as up-regulate melanoma antigens, led to the proposed triplet.

To test this notion, 514 patients were randomized in 1:1 fashion to either AZO/VEM/COB, or placebo/VEM/COB to determine if the immunotherapy had anything to contribute to this standard-of-care. (AZO had already shown some single-agent activity in this setting.)

All patients received a one-month run in of both small molecule inhibitors, followed by a placebo controlled combination of COM/VEM, plus or minus AZO. The primary endpoint was time to progression-free survival (PFS). 

“Interestingly, at six months after randomization there was no difference in investigator assessed PFS in the two arms,” McArthur reported. “However, by 12 months 54% of patients were free from progression in the Atezo arm vs 45.1% in the control arm. These differences between the two arms were at least maintained at 18 months and beyond, leading to overall PFS of 15.1 months vs. 10.6 months (p=0.024).”

Regarding the secondary endpoint, “There was no difference in overall survival at 12 months, however, at 24 months overall survival favored the Atezo arm, although this was not statistically significant,” said McArthur, adding that, “ongoing follow-up is critical to determine if these curves will further separate with time.”

Regarding the safety of this proposed regimen, higher elevations of liver enzymes were observed for the triplet, as well as more hypothyroidism, but these events were tolerable and in line with what was expected for anti-PD-L1 agents. 

The full report on the ImSpire150 trial is currently in press at The Lancet.

Commenting on the trial was plenary session moderator, and AACR president-elect, Antoni Ribas, MD, PhD: “(IMspire150) demonstrated that the triple therapy of two targeted drugs and an anti-PDL1 antibody was better than two targeted drugs in patients with BRAF positive melanoma. This is a practice-changing study for this indication.”

Also Ran

Closing out the trio of consecutively presented IO combination trials was IMbassador250, a phase III investigation of the antiandrogen drug, enzalutamide, plus or minus atezolizmab in patients with metastatic, castration-resistant prostate cancer (mCRPC; N=759). The primary endpoint was overall survival.

Observations from preclinical work, as well as intriguing results from a small handful of clinical data with these agents in this setting suggested that this combination was a good idea; the results from IMbassador250, as reported by Christopher Sweeney, MD, of Dana Farber Cancer Center, Boston, strongly suggested otherwise. 

“Twenty-three months after the first patient was enrolled, 300 deaths occurred,” said Sweeney. “The independent monitoring committee recommended stopping the study at that point due to futility. Specifically, there was less than one in 10,000 chance the study would show a clear improvement in overall survival.” 

So what happened?

According to Pam Sharma, MD, PhD, immunotherapy expert from the MD Anderson Cancer Center in Houston, there were many reasons why this trial may have failed. “One, is that prostate cancer has very few T cells, therefore, targeting PD-1 in a microenvironment where there are few T cells expressing these pathways may not lead to responses,” she began. 

Two, multiple immunosuppressive pathways are known to be expressed in this setting. Three, there are so few mutations in prostate cancer that effector T cells may not be able to recognize an adequate number of antigens that could lead to an anti-tumor response. 

Four, “The impact of enzalutamide on the immune response in the tumor microenvironment remains unknown. And five, these patients have had multiple prior therapies, leaving them with a poor immune response.”

 Or six: all of the above. 

TIL on Track

Tumor infiltrating lymphocytes, or TIL, are the immune cells – mostly T cells – that are found in the tumor tissue of many cancer types. Pioneering work done at the NCI demonstrated the efficacy of using TIL as a therapeutic, particularly in advanced melanoma.

This work is now being carried forward by industry, with refinements in TIL manufacture protocols, and investigations in many tumor types.

Ben Creelan, MD, of the Moffitt Cancer Center in Tampa, reported on the use of TIL, plus or minus the checkpoint inhibitor, nivolumab, in a phase I trial of checkpoint naïve patients with stage IV, non-small cell lung cancer (NSCLC). “The rationale for choosing PD-1 naive patients was to avoid potentially anergic or exhausted T cells harvested in a PD-1 failure setting,” he explained. 

For this study, 20 NSCLC patients first had TIL harvested from an accessible lesion. Following this, all patients received four doses of NIVO (standard of care) while serial CT scans were performed. NIVO responders continued on therapy; non-responders received TIL (up to 100 billion cells), with low-dose IL-2 (N=16). 

Results were encouraging, but modest. “Two patients have complete response by RECIST, going out past one year,” said Creelan. “In addition, one patient had a partial response (PR), and one patient has an ongoing unconfirmed partial response.” 

Observed adverse events were mild and were associated with IL-2.

Modest results aside, it was a significant proof of principle. “We are excited about carrying TIL further in lung cancer” said Creelan. 

A New Checkpoint

Ever since the approval of the anti-CTLA-4 agent, Yervoy, drug developers have been scouring the molecular landscape for other checkpoint inhibitors. Anti-PD-1 soon followed, then anti-PD-L1, then… a number of others that have failed, or have yet to prove themselves – despite considerable promise. 

Enter, COM701, a drug that targets something called PVRIG, a novel immune checkpoint target candidate discovered with bioinformatics by Compugen. “When looking for potential new targets, we tend to look at cell surface receptors, like PVRIG which are involved in T effector, or NK cell regulation” explained Ryan Sullivan, MD, of Harvard University, as he presented data on the use of COM701 in a phase I trial.

“Remarkable advances in tumor immunotherapy have been made in the last several years,” said Sullivan. “However, a number of unmet needs remain, particularly, finding better therapy for patients who are primarily refractory to, develop resistance on, or relapse following immune checkpoint inhibitors.” 

Sullivan reported the preliminary safety and anti-tumor activity of the proposed rescue, COM701, as monotherapy (n=18), or in combination with nivolumab in patients with advanced malignancies (n=13). In both monotherapy, and combination cohorts, COM701 was given in a dose-escalating, 3+3 design, and tumor types included NSCLC, ovarian, breast, endometrial, and colorectal.

Results so far for this ongoing trial show that, all patients have experienced adverse events – though none were considered serious, and related to that observation, a Q4 weekly dosing for the drug has been established.  

As for efficacy, “There was one partial response in arm A (mono) and one arm B (combo),” said Sullivan. Whereas the remainder of the patients mostly had stable disease. Of the partial responders, “these were both in patients with diseases predicted to not respond to immune checkpoint inhibition. Overall, the disease control was 69% for mono, and 75% for combo in diverse tumor types.” 

With these encouraging preliminary signals, the development of a potential new checkpoint inhibitor will move forward.

And that’s just half the story. 

A second installment of the conference, AACR Virtual Annual Meeting II, which will focus more on translational, rather than clinical research, is slated for June 22-24, 2020.


Post a Comment

OBR Archives

To view previous issues of OBR green you can visit our archives. The entire library of OBR green articles is searchable.