CME
Physicians: Maximum of 1.00 AMA PRA Category 1 Credit™
Released: August 31, 2022
Expiration: August 30, 2023
Christopher H. Lieu, MD:
At the 2022 ASCO annual meeting, several practice‑changing studies were presented for GI cancers. I would like to begin our discussion by highlighting 2 important presentations of data surrounding treatment for patients with mCRC.
Christopher H. Lieu, MD:
Several large phase III studies have been designed with the goal of determining the most appropriate first-line therapy for patients with mCRC with specific characteristics, including specific primary tumor location and molecular profile. These have included the CALGB SWOG 80405 and FIRE‑3 studies, both of which assessed chemotherapy plus cetuximab (an anti-EGFR therapy) vs chemotherapy plus bevacizumab (an anti-VEGF therapy) in the setting of first-line therapy for mCRC with wild-type RAS.4,5 In these studies, particularly after retrospective analyses, it appeared that outcomes with chemotherapy plus anti‑EGFR therapies were best for patients with wild‑type RAS mCRC with left‑sided primary tumors. To further address this question, the PARADIGM study was undertaken.
PARADIGM was a randomized, open‑label phase III study in which 823 patients with wild‑type RAS mCRC who had received no previous chemotherapy were randomized to receive FOLFOX plus panitumumab or FOLFOX plus bevacizumab.1 The primary endpoint of this study was OS in the left‑sided primary tumor population, with OS in the overall population as a coprimary endpoint. Secondary endpoints included progression-free survival (PFS), response rate, duration of response (DoR), and R0 resection rate.
Christopher H. Lieu, MD:
OS was improved in the left‑sided primary tumor population for patients receiving FOLFOX plus panitumumab vs FOLFOX plus bevacizumab (median OS: 37.9 vs 34.3 months; HR: 0.82; P =.031).1 Median OS was also improved in the overall population with FOLFOX plus panitumumab (36.2 vs 31.3 months with FOLFOX plus bevacizumab).
Christopher H. Lieu, MD:
The median PFS was not significantly different between the FOLFOX plus panitumumab and FOLFOX plus bevacizumab treatment groups, either in the left-sided primary tumor (13.7 vs 13.2 months) or overall populations (12.9 vs 12.0 months).1
Christopher H. Lieu, MD:
Response rates in the left‑sided population were higher among patients receiving FOLFOX plus panitumumab compared with those receiving FOLFOX plus bevacizumab (80.2% vs 68.6%).1
In the left-sided population, there was also a slight improvement in overall DoR among patients receiving FOLFOX plus panitumumab compared with those receiving FOLFOX plus bevacizumab (median DoR: 13.1 vs 11.2 months).
Christopher H. Lieu, MD:
Adverse events were consistent across both treatment groups, with expected adverse events associated with panitumumab and bevacizumab seen in the respective arms.1
Christopher H. Lieu, MD:
PARADIGM showed that in patients with wild-type RAS, left‑sided mCRC, the optimal treatment was chemotherapy plus panitumumab. Although this conclusion is consistent with those from previous studies, this was the first prospective study to make this conclusion.
I think that the response rate of 80% in the chemotherapy and anti‑EGFR therapy arm is among the highest that we have ever seen in prospective studies investigating first-line therapy for mCRC. This response rate in this study population is quite significant and should be considered when selecting therapy.
The key takeaway from this study is that to maximize outcomes, patients with wild-type RAS, left‑sided mCRC should receive chemotherapy and anti‑EGFR therapy as first-line therapy.
Rachna Shroff, MD:
This study was a proof of principle of what has been suspected for some time. I think that this study provides further support for acquiring detailed molecular profiling for patients with colorectal cancer (CRC) and understanding how to use the profile to provide the optimal targeted therapy.
These data also contribute to the ongoing discussion about sequencing of therapy. For example, after FOLFOX plus panitumumab, should patients be switched directly to a bevacizumab regimen? In my mind, these are the key questions.
Let’s return to one of the questions from earlier.
Christopher H. Lieu, MD:
CAIRO5 was a prospective, randomized phase III study comparing active systemic induction regimens in patients with mCRC and previously untreated, unresectable liver-only metastases. This analysis focused on the 291 patients with RAS/BRAFV600E mutated and/or right-sided primary tumors. Traditionally, these patients are not treated with anti-EGFR therapy, especially in the first-line setting.6
Patients were randomized to receive FOLFOX or FOLFIRI plus bevacizumab or the triplet chemotherapy regimen FOLFOXIRI plus bevacizumab. Patients received up to 12 cycles of these therapies and then switched to a maintenance regimen of 5‑fluorouracil plus bevacizumab. The primary endpoint of this study was PFS. Secondary endpoints included OS, ORR, R0/1 resection rates, and postoperative morbidity.
Every patient was evaluated by a central liver expert panel, which determined the surgical resectability of the disease. The panel included 15 liver surgeons and 3 abdominal radiologists.
Christopher H. Lieu, MD:
The median PFS was improved in the FOLFOXIRI plus bevacizumab group compared with the FOLFOX or FOLFIRI plus bevacizumab group (10.6 months vs 9.0 months).6 The data for OS were not yet mature at the time of the presentation.
The ORR was also significantly higher with FOLFOXIRI plus bevacizumab compared with the FOLFOX or FOLFIRI plus bevacizumab (53.5% vs 33.3%).
Christopher H. Lieu, MD:
More patients receiving the FOLFOXIRI plus bevacizumab triplet regimen experienced grade 3 adverse events compared with those receiving the FOLFOX or FOLFIRI plus bevacizumab doublet regimen, which is expected given the higher intensity of chemotherapy.6
Christopher H. Lieu, MD:
The percentage of patients undergoing resection was slightly higher in the FOLFOXIRI plus bevacizumab group (57%) compared with the FOLFOX or FOLFIRI plus bevacizumab group (46%), although the difference was not statistically significant.6
Among patients who underwent successful local treatment of their liver metastases, the median PFS was slightly higher in the FOLFOXIRI plus bevacizumab group compared with the FOLFOX or FOLFIRI plus bevacizumab group (12.7 months vs 11.9 months).
Christopher H. Lieu, MD:
In patients with initially unresectable colorectal liver metastases and right‑sided and/or RAS or BRAF‑mutated primary tumors, triplet chemotherapy with FOLFOXIRI plus bevacizumab was associated with improved efficacy compared with doublet chemotherapy plus bevacizumab. As expected, the triplet chemotherapy plus bevacizumab regimen was also associated with increased toxicity.
Additionally, the study demonstrated the feasibility of using a liver expert panel to increase the number of patients who were eligible for liver‑directed therapy.
Rachna Shroff, MD:
This study provides additional support for the dose intensification strategy of using FOLFOXIRI for appropriate patients with mCRC. One question I had is that although the median PFS was statistically significant, the HR is as high as 0.99. From the perspective of a medical oncologist treating these patients, we view survival endpoints—namely the PFS and OS—as being very important. However, similar to the PARADIGM study that we just discussed, the ORR was remarkable in terms of the difference between the triplet and doublet treatment regimens, with an improvement of 20%.
Christopher H. Lieu, MD:
I agree. I think that the key takeaway in this patient population is that the triplet therapy regimen is the preferred choice to maximize response. We often think about maximizing response in terms of converting patients from having unresectable disease to having resectable disease. Using this criterium, I think the triplet regimen would be the preferred choice for patients with these primary tumor characteristics and initially unresectable liver metastases, for whom response is critical.
Obviously, another consideration in treatment selection is whether a patient is an appropriate candidate for triplet chemotherapy. Patients need to be relatively fit to receive this therapy, and in patients with limited performance status, it is particularly important to consider whether they are able to withstand the rigors of triplet chemotherapy.
Overall, from the cumulative data presented at ASCO, the benefit of triplet chemotherapy can be observed mainly in right‑sided tumors vs left‑sided tumors. With left‑sided tumors, it appears that doublet chemotherapy is approximately as efficacious as triplet chemotherapy.
Christopher H. Lieu, MD:
The traditional neoadjuvant treatment approach to locally advanced rectal cancer includes CRT followed by surgery.2,7 More recently, HCPs have been utilizing a total neoadjuvant therapy approach that incorporates systemic chemotherapy prior to surgical resection. This approach can result in up to 35% of patients achieving a pathologic complete response.2
A small subset of patients with rectal cancer—approximately 5% to 10%—have dMMR and/or MSI-H disease.2 In patients with dMMR/MSI-H metastatic disease, immune checkpoint blockade as first-line and later-line therapy has shown significant efficacy.2,8,9
The current analysis was a single-arm phase II study examining the PD-1 immune checkpoint inhibitor dostarlimab as curative-intent therapy for patients with locally advanced (stage II or III) dMMR or MSI-H rectal cancer. Patients received 500 mg of dostarlimab every 3 weeks for a total of 9 cycles. After 9 treatment cycles, patients who achieved a cCR underwent a nonoperative surveillance period and those who had residual disease received CRT and, potentially, surgery. The primary endpoint assessed in this analysis was overall response to dostarlimab with or without CRT. Secondary endpoints included safety and tolerability.7
Christopher H. Lieu, MD:
Clinical response data were available for 14 of the 18 patients enrolled in this study. Of those who completed dostarlimab treatment, 100% (14/14) achieved a cCR, which was an incredibly high response rate.7 Dostarlimab was also very well tolerated.
Ultimately, these patients with dMMR or MSI-H locally advanced rectal cancer avoided chemotherapy, radiation, and surgery. In addition, at the time of ongoing follow-up, none of the patients had experienced a recurrence.
Christopher H. Lieu, MD:
This study clearly demonstrates the efficacy of immunotherapy in earlier stages of dMMR or MSI-H locally advanced rectal cancer.
Rachna Shroff, MD:
It’s hard to argue with these incredible results. The role of immunotherapy in advanced MSI-H rectal cancer has been demonstrated in multiple studies and there is an ongoing discussion about using immunotherapy at earlier cancer stages. In addition to these data, the NICHE trial also reported impressive results with neoadjuvant nivolumab plus ipilimumab in patients with resectable colon cancer, with pathologic response rates of 100% and 30% in patients with dMMR (n = 32) and proficient MMR (n = 30) tumors, respectively.10
What stands out to me is that, at least for now, this treatment allows patients to avoid surgery. Locally advanced rectal cancer can be quite morbid and significantly affect a patient’s quality of life. Although these results were from only a small cohort of patients, I think that if, as suggested by these and other results, immunotherapy provides durable responses, it may be absolutely life‑changing for patients with locally advanced rectal cancer.
Christopher H. Lieu, MD:
Yes, it’s hard to recommend a standard-of-care change based on small subsets. However, I think the data from this study are so outstanding that immunotherapy should be considered for patients with dMMR and/or MSI-H locally advanced rectal cancer. It’s worth noting again that these patients represent only a small subset (5%-10%) of patients with locally advanced rectal cancer. However, immunotherapy is a very promising treatment option for this group. Ongoing trials with immune checkpoint inhibitors in this space will continue to shed light on this topic; the EA2201 study (NCT04751370) of neoadjuvant nivolumab plus ipilimumab and short course radiation in MSI-H/dMMR rectal tumors will be of particular interest.
Let’s return to another of our questions.
Rachna Shroff, MD:
In patients with locally advanced pancreatic cancer, chemotherapy is the backbone of treatment.11 Less clear is the role of combining systemic chemotherapy with CRT and whether this provides a survival benefit. Although data from the LAP07 study suggested that there was no OS benefit to adding radiotherapy to chemotherapy for this patient population, the relevance of this trial has been questioned because it did not use modern chemotherapy.12
CONKO-007 was a randomized, phase III trial examining the role of CRT in patients with nonresectable locally advanced pancreatic cancer.13 Patients received gemcitabine or FOLFIRINOX induction therapy at the investigator’s discretion before being randomized to chemotherapy consisting of gemcitabine (1000 mg/m2/day on Days 1, 8, 15, 29, 36, 43, 57, 64, and 71) or FOLFIRINOX (Days 1, 15, 29, 42, 57, and 71) or a chemotherapy plus CRT combination consisting of irradiation (28 x 1.8 Gy) plus gemcitabine (300 mg/m2/day on Days 1, 8, 15, 22, and 29 followed by 1000 mg/m2/day on Days 57, 64, and 71). The primary endpoint was R0 resection rate. Secondary endpoints were OS, disease-free survival, resection rates, and survival following resection.13
Two important points to note are that the study used gemcitabine alone as a chemotherapy option vs the more contemporary regimen of gemcitabine plus nab‑paclitaxel, and that the primary endpoint was changed from OS to R0 resection rate due to insufficient recruitment.
Rachna Shroff, MD:
Just more than 50% of patients in each treatment group had T4 staging, making it an advanced patient population. Also, similarly to other studies examining the role of CRT in locally advanced pancreatic cancer, most patients had pancreatic head tumors.13
Rachna Shroff, MD:
At a median follow-up of 55.13 months in all randomized patients, the resection rate was just over 35% for both treatment groups. This means that approximately one third of patients in each treatment group were converted to resectability.13
Although not significantly different, the R0 resection rate was numerically higher in the chemotherapy plus CRT group compared with the chemotherapy group (25% vs 18%, P = .1126).
However, examination of the important circumferential resection margin (CRM) revealed that positive CRM rates were lower in the chemotherapy plus CRT group compared with the chemotherapy group (7% vs 16%).
Rachna Shroff, MD:
Similar to the results in all randomized patients, in the subset of 122 patients who underwent surgery, the R0 resection rate was improved in the chemotherapy plus CRT group compared with the chemotherapy group (69% vs 50%, P =.0418) and the positive CRM rates were lower in the chemotherapy plus CRT group compared with the chemotherapy group (18% vs 45%).13
Rachna Shroff, MD:
Importantly, as in other studies, the addition of CRT to chemotherapy did not lead to a PFS or OS benefit compared with chemotherapy alone.13
Rachna Shroff, MD:
In a subgroup analysis, the OS was improved among patients who underwent surgery compared with those who did not (median OS: 19 vs 14 months, respectively). This result confirms what was already known—that patients who undergo surgery seem to have overall better biology and/or survival outcomes compared with those who do not undergo surgery.13
Another subgroup analysis examined the OS among 112 surgical patients who were treated with FOLFIRINOX. In this subgroup, the OS was similar between the group that received FOLFIRINOX followed by chemotherapy plus CRT and the group that received FOLFIRINOX followed by chemotherapy (median OS: 22 vs 21 months, respectively). This suggests that modern chemotherapy supplants the need for radiation.
Rachna Shroff, MD:
Not surprisingly, the grade 3 toxicity rate was higher in the group receiving chemotherapy plus CRT compared with the group receiving chemotherapy (73% vs 39%). Of note, the grade 4 toxicity rate was also slightly higher among patients in the chemotherapy plus CRT compared with those in the chemotherapy group (6% vs 1%).13
Rachna Shroff, MD:
The CONKO-007 study showed that the addition of CRT after induction chemotherapy improved the R0 resection rate among the subgroup of patients who underwent surgery, but not in the intention-to-treat population.13 In the intention-to-treat population and the subgroup of 122 patients who underwent surgery after receiving randomized treatment, the addition of CRT to chemotherapy significantly reduced the R1 resection rate compared with chemotherapy alone. However, PFS and OS were similar in both treatment groups in the intention-to-treat population.
The overall conclusion was that induction chemotherapy followed by the addition of CRT and surgery was feasible and, in a subset of patients, resulted in a 5‑year OS rate of 26.9%.
Although there is potentially a role for CRT in treating patients with unresectable locally advanced pancreatic cancer, I am beginning to question the absolute need for it with the availability of FOLFIRINOX and gemcitabine plus nab‑paclitaxel. The PFS and OS data are the most meaningful endpoints here. If adding radiation increases the negative aspects of treatment—such as increasing grade 3 toxicities—without dramatically improving OS, it makes me question the role of CRT.
The results of this study contribute to an accumulating body of information being used to address this ongoing question.
Christopher H. Lieu, MD:
I really like your point that questions whether we should be using treatment modalities that do not improve PFS or OS. Especially with pancreatic cancer, there probably is a patient population that is deriving benefit from CRT, but we do not know what that population is. Therefore, offering CRT to all patients without there being a clear survival benefit may not be the best therapeutic strategy.
Rachna Shroff, MD:
I think it’s important to note that, to date, the best OS data in resectable disease are associated with adjuvant FOLFIRINOX treatment, which suggests that chemotherapy options are improving. We are still trying to understand the best approach to perioperative and neoadjuvant treatment. One strategy is preoperative FOLFIRINOX or gemcitabine plus nab‑paclitaxel. I think that as we improve our chemotherapy options, there could be further arguments against using CRT in all patients with localized pancreatic cancer. The questions remains, which subset of patients benefits most.
Rachna Shroff, MD:
Treatment options for BTCs are extremely limited. Options for advanced BTCs include conventional gemcitabine and cisplatin‑based regimens with or without durvalumab.14,15 However, BTCs are molecularly complex and there are numerous targets for which targeted therapies are approved or in development.
One potential target is HER2 overexpression or gene amplification, which has been reported in approximately 14% to 15% of BTCs.16 HER2 overexpression/gene amplification is primarily observed in extrahepatic cholangiocarcinomas (CCAs) and gallbladder cancers, and is seen more rarely in intrahepatic CCAs. The MyPathway study demonstrated the potential efficacy of targeting the HER2 pathway in a highly refractory BTC cohort that was treated with trastuzumab plus pertuzumab.17
T-DXd is an anti‑HER2 antibody–drug conjugate that has demonstrated remarkable efficacy in breast cancer—even among the HER2‑low patient population. It has also demonstrated survival benefits in GI malignancies, and is now approved for patients with locally advanced or metastatic HER2-positive gastric or gastroesophageal junction adenocarcinoma who have received a previous trastuzumab-based regimen.18-20
HERB was a multicenter, single-arm phase II study examining T-DXd in 30 patients with unresectable or recurrent HER2-expressing BTCs.3 HERB was part of the Japanese SCRUM effort, which is examining precision oncology and real-time molecular profiling. In this study, immunohistochemistry (IHC) and in situ hybridization (ISH) were used to categorize patients with BTCs into 1 of 3 HER2 categories during screening: HER2 positive (IHC 3+ or IHC 2+/ISH positive), HER2 low (IHC/ISH 0/+, 1+/–, 1+/+, or 2+/–), or HER2 negative (IHC/ISH 0/–). Patients with HER2-expressing (HER2-positive or HER2-low) BTCs who were refractory or intolerant to gemcitabine received 5.4 mg/kg IV T-DXd every 3 weeks until disease progression, unacceptable toxicity, or withdrawal. The primary endpoint was a confirmed ORR by blinded independent central review (BICR) in HER2-positive patients. Secondary endpoints were OR by local investigator review, DCR, PFS, OS, and safety. Efficacy in the HER2-low population was an exploratory component of the study.
Rachna Shroff, MD:
Among the 30 enrolled patients, 22 had HER2-positive disease and 8 had HER2-low disease, making it a much smaller, exploratory cohort.3 Among HER2-positive patients, 50% had gallbladder cancer and 27.3% had extrahepatic CCA. The specific BTCs represented in the HER2-low population were more evenly divided across cancer type.
Most HER2-positive and HER2-low patients (72.2% and 62.5%, respectively) were highly refractory, having received at least 2 previous lines of therapy.
Rachna Shroff, MD:
Among the 22 HER2-positive patients, T-DXd was associated with an impressive 36.4% confirmed ORR by BICR and an 81% DCR.3 Of note, in 8 patients with HER2-low disease, the ORR was 12.5%, with a 75% DCR.
Rachna Shroff, MD:
HER2-positive patients experienced an equally impressive and similar 36.4% confirmed ORR by investigator review.3 As in the BICR analysis, an ORR of 12.5% was observed in patients with HER2-low disease.
Rachna Shroff, MD:
At a median follow-up of 7.1 months, the median PFS was 5.1 months and the 6‑month PFS rate was 40.9% among patients in the HER2-positive population.3 The median PFS was lower (3.5 months) among the HER2‑low population, although it is important to remember that only 8 patients were included in this population.
Rachna Shroff, MD:
The median OS was 7.1 months in the HER2-positive population and 8.9 months in the HER2-low population.3
Rachna Shroff, MD:
We know from previous studies that T-DXd can be associated with treatment‑emergent adverse events of interest that include cytopenias and ILD/pneumonitis.
Among the 32 patients in the safety population, grade 3 or greater anemia, neutropenia, and leukopenia occurred in 53%, 31.5%, and 31.5% of patients, respectively.
Any‑grade ILD/pneumonitis was observed in 25% of patients and grade 3 or greater ILD/pneumonitis occurred in 12.5% of patients. In other words, grade 3 or greater ILD/pneumonitis developed in 50% of patients who experienced this event. Development of ILD/pneumonitis was not related to the previous number of treatments or HER2 status. However, 37.5% of the patients who developed ILD/pneumonitis had lung metastases and 37.5% had a history of smoking.
Rachna Shroff, MD:
Oncologists have been awaiting data with T-DXd in BTCs, and the ORR seen in the HER2-positive population is intriguing and merits further exploration. Even more interesting are the results from the HER2‑low population, because a fair number of patients with BTC fall into this category. If T-DXd is effective in this population, as well as the HER2-positive population, it would be an incredible treatment option for patients.
Having said that, it is important to be mindful of the association of T-DXd with ILD/pneumonitis. One consideration is that all patients with BTC likely received systemic therapy with gemcitabine, which is known to be associated with ILD/pneumonitis. Therefore, it is possible that gemcitabine treatment primes patients for developing ILD/pneumonitis. The rates of ILD/pneumonitis observed in this study were higher than those observed in some other studies with T-DXd. Therefore, to develop a better understanding of the relationship of T-DXd to ILD/pneumonitis, it will be important to follow ILD/pneumonitis rates in a larger study.
Christopher H. Lieu, MD:
These data add to emerging evidence supporting the role of molecularly directed therapy for BTCs, which is exciting. Results from the MyPathway study showed that HER2 is an intriguing target for therapy; the response rate observed in this study for a very difficult-to-treat population is very exciting. The data showing that the HER2‑low population had any response to treatment is also exciting because it is something that is not usually seen in HER2‑amplified mCRC.
The HERB study is a really exciting start for examining T-DXd in BTCs, and the oncology community will be looking forward to seeing additional data in larger patient populations.
Christopher H. Lieu, MD:
Data from these additional studies of interest were also presented at ASCO 2022:
Back
