Later-Line Therapy in Metastatic CRC Calls for Safety Considerations in Refined Decision Making

Kanwal P.S. Raghav, MBBS, MD

Given the modest responses associated with regorafenib (Stivarga) and trifluridine/tipiracil (TAS-102; Lonsurf) in metastatic colorectal cancer (mCRC), combination strategies with these therapies plus TKIs or checkpoint inhibitors are being investigated in patients with mismatch repair proficient [pMMR] tumors, explained Kanwal P.S. Raghav, MBBS, MD.

Moreover, Raghav highlighted that molecular profiling is critical to identify patients with certain subtypes of mCRC, guide treatment decisions, and optimize available therapies prior to third- or fourth-line options.

“These are essential discussions in the treatment of patients with mCRC because by the time most patients reach [the salvage-line setting], they have a larger burden of disease and may be sicker. Therefore, appropriate treatment selection is important to improve their outcomes and preserve their quality of life [QOL],” said Raghav, an associate professor in the Department of Gastrointestinal (GI) Medical Oncology in the Division of Cancer Medicine, and medical director of the Division of Ambulatory Treatment Centers at The University of Texas MD Anderson Cancer Center, in an interview with OncLive^® during an Institutional Perspectives in Cancer (IPC) webinar on gastrointestinal malignancies.

The virtual meeting covered the evolving treatment landscape of mCRC, emerging options in hepatocellular carcinoma (HCC), and significant updates in gastroesophageal cancer.

In the interview, Raghav, who chaired the event, discussed nuances to treatment selection in mCRC, the significance of the phase 2 ReDOS trial (NCT02368886) with regorafenib, and how the paradigms for mCRC, HCC, and gastroesophageal cancers have expanded from areas of unmet need to ones with more robust armamentariums.

OncLive^®: How do you approach treatment selection for patients with mCRC? Should molecular testing be considered in all cases?

Raghav: [Patients with] mCRC [are] treated with a wide variety of agents, and [treatment selection] is based on molecular profiling and what subset of mCRC [a patient] has. In most cases, patients will get first-line treatment or second-line treatment and subsequently third-line treatment. With recent developments, the definition of these lines of treatment is getting blurred; however, most patients with mCRC will progress through their initial lines of therapy and reach a point of salvage therapy.

Broadly speaking, CRC [management] is a compendium of cytotoxic chemotherapy, some targeted agents, immunotherapy, and some salvage [therapies]. In a general sense, if we think about the workup of a patient with mCRC, I am a big advocate of up-front molecular profiling [because] there are 2 advantages. [Molecular profiling] changes the patient’s standard-of-care management in a big way. Also, we can think ahead in the future to prep these patients for clinical trials, especially if we can identify rare subsets of tumors.

Broadly, when we start with CRC treatment, the first decision [depends on whether patients have] microsatellite instability–high [MSI-H] or mismatch repair deficient [dMMR] tumors. If we have [a patient with] a dMMR tumor—which [comprise] 4% to 5% of all mCRC—the treatment is up-front immunotherapy. This [is based on] the [KEYNOTE-177 study (NCT02563002), which evaluated] pembrolizumab [Keytruda] vs chemotherapy. [The] study showed a significant improvement in overall survival [with pembrolizumab]. All of these patients [with dMMR CRC] should receive up-front immunotherapy [because] that is more effective and better tolerated [vs chemotherapy].

Once we have removed the MSI-H/dMMR subset, we move to the pMMR group, which is larger. Within this group, there are a couple of subsets, including RAS-mutant CRC, which [accounts for] about 50% to 55% of all CRC. That subset, unfortunately, has very limited treatment options, so they are essentially treated with cytotoxic agents and anti-VEGF agents, such as bevacizumab [Avastin]. In first-line treatment, we can discuss options regarding triplet and cytotoxic doublets depending on how [fit] the patient is.

Within the RAS-mutant subset is a smaller subset of KRAS G12C–mutant CRC that should be identified early. Agents and clinical trials are available to [treat] that patient subset.

The remaining KRAS wild-type patients are divided into 4 categories, including patients with RAS wild-type, BRAF-mutant CRC. Patients with a BRAF mutation go down the line of the BEACON CRC trial [NCT02928224] regimen of combination BRAF- and EGFR-directed [targeted] therapies for which there are clinical trials. Also, in the second-line setting, [this dual-targeted approach] is standard of care. Then, there is RAS wild-type, HER2-amplified disease, in which the options include dual anti-HER2 therapy with trastuzumab [Herceptin] plus pertuzumab [Perjeta], trastuzumab plus lapatinib [Tykerb], trastuzumab plus tucatinib [Tukysa], or clinical trials. There is also fam-trastuzumab deruxtecan-nxki [Enhertu], which is an anti-HER2 antibody-drug conjugate, that can be used for this subset. [HER2-amplified CRC represents] a small subset, but many agents are available for those patients.

After we have removed those 2 subsets, what remains of RAS wild-type CRC can be divided into left- or right-sided CRC. In left-sided CRC that is BRAF negative, HER2 nonamplified, and RAS wild-type, I would use an anti-EGFR agent, up front if possible or at least in the second-line setting. Right-sided RAS wild-type tumors do not respond well to an EGFR-directed agent, so I would try to move them toward clinical trials or adopt the anti-EGFR approach in a later line of treatment. Of course, after all of this is done, [treatment selection] comes down to the third-line setting.

What considerations guide treatment selection between available third-line options?

A few options are available in this salvage-line setting that are standard of care, including regorafenib and TAS-102. Other strategies are also adopted in the clinic, but they are not supported with the same level of robust evidence as with a randomized phase 3 trial. One of [these alternative approaches] is to rechallenge with chemotherapy.

The safety profiles of the drugs guide most of the selection. If we look at the clinical trials, [the results] showed that these drugs were similar in design; therefore, there is not a whole lot of variability [to guide treatment decisions]. However, if I have been giving chemotherapy to someone and bone marrow depression has been a major toxicity, I am not going to use TAS-102 because [the agent’s] bone marrow toxicity is limiting. If somebody had peritoneal surgery or diarrhea as a major issue on prior lines of treatment, regorafenib is not the best drug to use in that situation. Otherwise, the most important factor is to choose patients carefully and make sure [decisions are informed by] what the trials have shown.

These drugs are effective, but they have only been shown to be effective in patients with good performance and clinical status. Fortunately, we have some clinical trials running at our center. One of them is looking at circulating tumor DNA monitoring and early signs of progression in patients in the salvage-line setting. We usually put [eligible] patients on that trial, in which regorafenib is generally used before TAS-102.

There are data about TAS-102 in a regorafenib-treated population, but no data exist [in the opposite sequence]. Most of these patients are coming through the first and second line of chemotherapy, so putting regorafenib in between [chemotherapy and TAS-102] changes the mechanism of action. Essentially, by the time we come [to the salvage-line setting], our primary emphasis is to encourage participation on clinical trials.

How are you incorporating the findings from the ReDOS trial into your approach to treatment with regorafenib? Is the goal of treatment to reach the maximum-tolerated dose or should dosing be individualized?

Even before the ReDOS trial, we were using a more dose-escalation–based strategy [with regorafenib] because most patients do not tolerate [the drug at] a very high dose. ReDOS also showed us that we probably don’t need [to treat] with the highest dose [of regorafenib], but rather] we need patients to be on [regorafenib] for a longer period. If we can achieve that by a dose-escalation strategy, we are more likely to keep patients on treatment.

Also, when we look at the phase 1 [trial] of this drug, the dynamics of 120 mg are very equivalent to those of 160 mg. We try to dose escalate as quickly as possible to the highest dose level possible, provided that patients maintain a good QOL and their [toxicity] is well managed. That is our standard practice across the board.

What agents could be combined with regorafenib or TAS-102 to potentially improve upon the modest response rates observed in the later-line setting?

Clinical trials are ongoing that are combining [regorafenib or TAS-102] with other drugs. There is also great emphasis on trying to get immunotherapy into the landscape for mismatch repair proficient disease. Other ongoing studies are evaluating a combination of TKIs like regorafenib, lenvatinib [Lenvima], or cabozantinib [Cabometyx], along with anti–PD-1 drugs because of the immunomodulatory effect these drugs can have to increase responses to PD-1 inhibitors. As single agents, there is no response [to PD-1 inhibitors].

Those are promising trials. The data from the phase 2 LEAP-005 trial [NCT03797326] with lenvatinib plus pembrolizumab were presented, [showing] a response rate of 20% in this population, but we have to see whether this will pan out in a larger study.

The bottom line is that, yes, these drugs have given us options for our patients. However, these are not curative options, and the clinical benefit is modest. After first-line chemotherapy, clinical benefit is modest for nearly all treatments apart from select targeted therapies. Therefore, as soon as we diagnose a patient with mCRC, we should move them toward clinical trials.

Where do you see a role for rechallenging with chemotherapy?

Whether to rechallenge with chemotherapy is an important question, so I want everyone to think about rechallenging and to be able to differentiate between rechallenging and reintroducing. Sometimes we do things because we want to do things, but patients can get into trouble with toxicities. Whenever we are selecting therapies, [we want to] have a frank and candid discussion about the risk-benefit profile of these salvage-line options, including rechallenging with chemotherapy.

During the IPC meeting, Ahmed O. Kaseb, MD, of The University of Texas MD Anderson Cancer Center, discussed emerging options in HCC. What have been the most significant advances in this field?

For a very long period, HCC was a graveyard for treatments. We basically only had sorafenib [Nexavar] for a very long time with a slew of negative trials. Suddenly, we have seen such a big boom in HCC that it has become a complicated treatment landscape.

Broadly speaking, immunotherapy and next-generation TKIs have taken HCC by stride. The success of lenvatinib, regorafenib, and cabozantinib, followed by promising responses of single-agent immunotherapy and better promises of dual checkpoint inhibition, [is exciting]. Atezolizumab [Tecentriq] plus bevacizumab [has been integrated] as a successful first-line treatment option. There are emerging data with other similar TKI/immunotherapy combinations. This is such a broad and evolving field, and patients have a lot of options.

Finally, Mariela Blum Murphy, MD, of The University of Texas MD Anderson Cancer Center, spoke about updates in gastroesophageal cancer during her presentation at the IPC webinar. What were your key takeaways from her lecture?

Pretty much everything I’ve said about HCC can be said about gastroesophageal cancer. Unfortunately, [gastroesophageal cancer is] a [field with] great unmet need globally. There is not a very big tumor population in the United States, but globally, it is a very big burden of disease. Therefore, it is heartening to see improved responses and survival outcomes with the combination of immunotherapy [for] many subsets of gastric and gastroesophageal cancers. Of course, we must differentiate between histology and [perform] PD-L1 testing. It is still good to see that we are moving the needle forward, although not as much as we would like.

As for HCC and CRC, as well as gastroesophageal cancer, the reasons why this has been successful is because of clinical trials. Therefore, please encourage patients for clinical trial participation as early as possible, not only when they have exhausted all lines of [available] therapy. We know that even though there have been strides, [many] patients with these diseases do not benefit with what we consider standard of care today.