Neeta Somaiah, MD

With the approval of 2 new drugs in 2020 for GIST [gastrointestinal stromal tumor], I think there is still a lot of room for improvement. A couple of avenues of improvement are to try to see if with increasing efficacy we can improve the toxicity profile for patients, especially once they switch from imatinib to second- and third-line therapies. With that there was an ongoing phase 3 study looking at ripretinib versus sunitinib to see how it would compare, in terms of efficacy and safety, to sunitinib.

The other big unmet need for GIST, and we hope there is more research done here, is for those patients with wild-type GIST, such as the patient we discussed. Even though we can cycle them through all the available lines of therapy, these patients’ responses to these TKIs [tyrosine kinase inhibitors] tend to be less compared to patients with 

There are some ongoing studies. There was one with Temodar for patients with wild-type or 

-mutated GIST, and we hope there will be more studies that will target that population as well.

The other biggest thing when you look at GIST progression is development of resistance. How can we tackle that, when patients respond so well to frontline therapies? Can we decrease the emergence of these resistant clones and help patients have a better outcome without having to cycle through so many drugs? For that patient, there are trials ongoing where patients have received a combination of imatinib with another drug. There was a MEK inhibitor, binimetinib, plus imatinib study, that was presented at ASCO 2020 [American Society of Clinical Oncology annual meeting]. These are early studies, and we’ll need more testing because though it looks promising in the single arm of combining imatinib with binimetinib, only a randomized study will show whether the efficacy improvement is more compared to imatinib alone, and does that justify the increased toxicity that you would see with a combination?

There are other things that are ongoing where you cycle drugs. There was a study of sunitinib alternating with regorafenib. Though you might be able to suppress different clones of resistance, there were challenges associated with such a treatment. Combining both the TKIs does lead to increased toxicity.

The other study that is available is a phase 3 study of crenolanib, which is also specific to the 

 D842V population. This study was ongoing even before avapritinib got approved, but now it probably will be used only after the treatment with avapritinib for patients with a subset of that resistant mutation.

With immunotherapy, there was initially a lot of interest in looking at checkpoint inhibitors in combination with the TKIs or alone. There still needs to be a lot of research done in that area because there are some case reports and some trials showing encouraging responses, but other trials are showing no response in patients with GIST. We still need to research that more.

This is an exciting year for the treatment of patients with GIST, with the approval of 2 new drugs. Avapritinib was approved for patients with exon 18 

-mutated GIST, which is a small subset, but these patients have been previously resistant to other TKIs, and the approval of ripretinib for patients in fourth-line GIST who have progressed, or were intolerant to imatinib, sunitinib, and regorafenib in the previous lines.

It’s important to recognize the availability of these drugs so that patients with GIST can have access to these agents that can have a significant impact on their survival.

A 68-Year-Old Man With Gastrointestinal Stromal Tumor

A 68-year-old man complains of a 4-month history of decreased appetite, vague abdominal discomfort and a sensation of fullness

PMH: DM, medically controlled; colonoscopy at age 50 was unremarkable; no family history of cancer

PE: abdominal pain on deep palpation; otherwise unremarkable



Labs: Hb 10.5 g/dL, plt 105 x 109/L; other lab values WNL

Endoscopy: showeda submucosal ~6 cm mass with ulceration

EUS: irregular borders on extraluminal surfaces with marginal halo and hyperechogenic spots

FNA biopsy: mitotic activity showed >5 mitoses/50 HPFs

Abdominal/pelvic CT confirmed a 6.2 cm lesion with indistinct margins in the body of the stomach

MRI showed 3 small hepatic lesions consistent with metastases

He was started on imatinib 400 mg PO qDay, continued for 24 months until he complained of increased abdominal pain and an additional 8-lb weight loss

Treatment initiated with sunitinib 50 mg PO qDay for 4 weeks, with a 2-week drug-free interval

Treatment was well tolerated for 3 cycles when he developed nausea and vomiting; dose reduced to 25 mg PO qDay without resolution of AEs; sunitinib was discontinued

Regorafenib 160 mg (four 40 mg tablets) PO qDay for the first 21 days of each 28-day cycle was started, poorly tolerated

Repeat lab work showed: Hb 8.9, AST 65 IU/I; regorafenib was discontinued

The patient was started on ripretinib 150 mg PO qDay

Videos

Future Directions of Gastrointestinal Stomal Tumors

Now I’m going to discuss data from the phase 3 INVICTUS trial, which led to the approval of ripretinib in fourth-line GIST [gastrointestinal stromal tumor]. The study was a randomized, multicenter study where 129 patients were randomized in a 2:1 fashion to either receive ripretinib, 150 mg once daily, or placebo.

Patients were imaged initially every 4 weeks, so we could pick up progression early, especially in the placebo arm.

Once they were unblinded at the time of progression, if they were on ripretinib, they had 3 choices: either to continue the same ripretinib dose if their physician thought they were deriving clinical benefit; or if they felt they could tolerate it, they were allowed to escalate the dose to 150 mg twice a day; or discontinue based on investigator or patient choice.

If you were on placebo, you had the option to cross over to ripretinib, 150 mg once daily. These patients were followed, and the majority of the patients on this study were heavily pretreated. There were some patients in the fourth-line setting, but many patients had received up to 7 prior lines of therapy.

Patients had a wide range of primary mutations, so they did look at the primary mutations on their archival specimens that were available. The study also involved the collection of liquid biopsies, though we don’t have that result yet in detail.

The majority of patients on this study did have a 

 mutations, a few of them were wild type, and for a few we didn’t know their mutational status in baseline.

The results of the study showed a statistically significant improvement in the progression-free survival for patients who received the ripretinib versus placebo, which was the primary end point. It was 6.3 months for patients on ripretinib versus 1 month for patients on placebo. At the first scan timepoint, most patients on placebo were progressing.

This is quite classic for patients with GIST. When you remove TKI [tyrosine kinase inhibitor] therapy, or especially beyond third line, these patients tend to do very poorly. A 6.3-month progression-free survival benefit on the ripretinib arm was quite impressive. The response rate showed that 9% of patients on ripretinib had a RECIST [Response Evaluation Criteria in Solid Tumors] response by blinded independent central radiology review, and 0% of patients on placebo had a response. This also translated to a survival advantage, where patients on the ripretinib arm had a median overall survival of 15.1 months compared to 6.6 months on placebo.

These data were very encouraging and led the FDA to approve this drug not too long ago for fourth-line and beyond patients. They did not require baseline mutational testing, so patients with all mutations in the fourth-line setting in the absence of any other agents are approved to receive ripretinib and are likely to derive benefit from it.

The other important thing to know with new drugs is their safety profile. What’s interesting is most patients who have received sunitinib and regorafenib in the second- and third-line setting, when they switched to ripretinib, most of them will notice an improvement in some of the symptoms, especially the hand-foot syndrome.

Patients with ripretinib do have palmar-plantar erythrodysesthesia [PPE], which is what we call the hand-foot syndrome, symptoms, but they’re usually much milder than what we see with sunitinib or regorafenib. Most people are good at managing adverse effects with the previous well-known TKIs. I usually tell patients to continue moisturizing their palms and soles well, and I’ve not had to do a dose reduction for PPE in patients with ripretinib, but just local management.

The other important thing to counsel patients about with ripretinib that was seen on the study, was a little over a 50% rate of alopecia. Unlike other TKIs, where with some of them you see a change in hair color, with ripretinib patients do end up losing their hair. Sometimes it’s hair thinning, but a lot of them will have hair loss.

I would say for most patients in the fourth-line setting, they usually don’t care so much about their hair loss, but for a lot of patients, and especially female patients, it is a big change, so it is important to counsel them up front about this adverse effect that they might experience, and give them adequate resources to manage it, whether it be a wig or what we call a cranial prosthesis, or advise them how to approach this.

You could also send them to a dermatologist, not only because of the alopecia, but there is another adverse effect that that should be monitored, and this is the development of skin lesions that rarely could be precancerous or cancerous. There was a slight increased incidence of the squamous cell cancers and even melanoma that was very minimal. For most of these patients, I do recommend following up with the dermatologist at the sign of any new lesions, and they should at least get a yearly skin examination by a dermatologist.

In addition, the other adverse effects were quite similar to what we would we see with imatinib. There is some fluid retention, some myalgias, but not much.

In general, the grade 3 or 4 toxicity profile is not very different from placebo. Unlike sunitinib and regorafenib, most patients at the starting dose of 150 mg once daily were able to tolerate the drug well, and only 7% needed a dose reduction. With this drug, I usually start at dose, and if they progress, I have dose escalated as well, and they have tolerated it well.

It’s important to keep track of these adverse effects and know them. Another one is high blood pressure, which we’re used to monitoring with sunitinib and regorafenib. It’s actually much less common with this, so sometimes you might need to reduce their antihypertensive dose that they’re on, but some patients might have an increase in blood pressure if they haven’t seen that with the previous drugs as well, so it needs to be monitored.

With any new TKI it’s important to follow laboratory test parameters, and I usually do them every 2 weeks initially for the first month or two on ripretinib, and then you can switch to every month thereafter. It’s important to follow for anemia, the liver function test, but the most common things that are seen with ripretinib are a decrease in phosphorous and an increase in lipase.

The clinical significance of these are not very well known because it hasn’t been linked to an increase in pancreatitis, but it’s important to follow and recognize that this can happen with ripretinib. Of course, for the phosphates, you replace it as you would with any other therapy, and then follow the patients to make sure they’re not developing any other concerning symptoms.

The other finding of importance that came to light with INVICTUS that we all knew, was patients on placebo don’t do very well. You do want to make sure patients with GIST who are progressing get exposed to as many effective or active lines of therapy as possible, and don’t get too much of a break between each of these TKIs because they could rapidly decline in their clinical and performance status.

Some of these data were available through the quality of life parameters that were collected on the INVICTUS study. At the virtual ASCO 2020 [American Society of Clinical Oncology annual meeting], 2 abstracts were presented. One showed that though patients with ripretinib did have a little over 50% rate of alopecia, and the palmar plantar erythrodysesthesia was a little over 20%, the effect of these adverse effects on their quality of life was minimal or decreased over time.

The other thing that was important is patients on ripretinib were able to maintain their quality of life metrics that were collected with 2 questionnaires looking at the visual analog scale and also looking for the role of their physical function in their daily activities. That showed that on ripretinib, there was slight improvement, the numbers were not significant there, but at least it was stable. The patients on placebo noted a decline in their functional status. It is important not to take too long of a break, and know that there is a fourth-line therapy available so patients can switch if they’re progressing on the third-line therapy.

It’s also important to recognize that the mechanism of action of ripretinib is a little different than the previous TKIs. This was a drug specifically designed to target these driver mutations, and it is what we call a switch control inhibitor. In patients with GIST, most of the secondary mutations will be seen in the ATP [adenosine triphosphate] binding site, which are the exon 13 and 14 mutations. Also, mutations are seen in the activation loop, which are exon 17 and 18 mutations. Some drugs effectively target one or the other.

Ripretinib can bind to the switch pocket, the ATP binding site, and also connect to the activation loop, and it locks the molecule in the inactive position. It’s a type 2 inhibitor that locks that in an inactive position, so it doesn’t let the activation loop swing back and cause the transcription for downstream factors in 

Understanding the INVICTUS Trial

Patients with GIST [gastrointestinal stromal tumor] who progress beyond frontline therapy on imatinib and move on to second-line therapies, invariably require third-line, fourth-line therapy, and beyond because their time on the TKIs [tyrosine kinase inhibitors] after imatinib tend to be more short-lived. This points to the type of resistance that they have developed in their tumors and might point to whether they get benefits on their second, third, or fourth-line therapies.

When they come to see me, a third-line approved agent is regorafenib, and most patients with an exon 17 secondary mutation, and even some patients with wild-type GIST, do quite well on regorafenib. They do invariably progress, and prior to recently, we didn’t have an approved fourth-line agent, and we used to treat them with other available TKIs that you could prescribe, such as pazopanib, nilotinib, and ponatinib. These are tough to get approvals for insurance, but there are published data of response, and on occasion, patients would respond to these TKIs.

The standard thing to do as well, if you don’t have an approved agent, is to put the patient back on the TKI they have previously responded to well because the progression tends to be slower when they’re on the TKI than when they’re off the therapy.

Progressing Gastrointestinal Stromal Tumor

After frontline therapy, what are the approved or available lines of therapy in the second line? Approved therapy for second-line treatment of GIST [gastrointestinal stromal tumor], it’s sunitinib, which was approved based on a randomized, placebo-controlled trial. When patients are progressing on imatinib, it’s important to recognize the types of progression.

Occasionally patients who’ve had a good response to imatinib, if they’ve had a 

 mutations, they’re responding well, they develop very cystic lesions, and then over time within these lesions you might start seeing what we call nodular progression, or small nodules of solid tumor that start appearing and slowly growing over time. What that signifies is that there is probably development of a resistance mutation in a little clone of cells that is now growing.

The second type of progression of course is when you have more global progression in multiple tumors or multiple metastases all at the same time, and it’s not nodular. These tend to be either primary resistance patterns, where the patient never responded to imatinib and continued to progress, or they were responding, but then multiple lesions started to progress.

The type of progression tends to point toward the type of resistance the patient might have to the treatment. If you have more global progression, it points to either a resistant mutation up front like the 

 D842V. Sometimes these patients might have NF1 [neurofibromatosis type 1], some of these wild-type tumors, some mutations that don’t respond to imatinib might have more of a global progression.

In patients who were responding and develop multiple areas that are progressing, it signifies secondary mutations that might have developed after the primary mutation. Nodular progression usually signifies a secondary mutation that has developed on top of the primary 

What are secondary mutations that are common? I said that 60% to 70% have a 

 primary mutation, and they’re most common in exon 11 and exon 9. Primary mutations in exon 13, 14, and 17 are very rare, it’s just 1%. When you look at secondary mutations, and this again, not everybody gets this evaluated because not everybody gets a biopsy at the time of progression, but if you do a biopsy of the progressing lesion, you might find in addition to their primary mutation that they now have a mutation in exon 13, or exon 17 and 18. These tend to be the more common secondary mutations.

The type of secondary mutation also points sometimes to whether you would respond to a certain second-line therapy. In the second-line treatment, sunitinib is the standard, unless you have an available clinical trial to offer in the second line. If you do have secondary mutation data, either from a tumor biopsy or perhaps from circulating tumor cells, what we call liquid biopsy, you could take that into consideration. Though I usually use caution; I like to follow the standard second-line and third-line treatment because sometimes biopsy of 1 tumor site might not represent the secondary mutations that might be present in all of the metastatic sites in the patient.

If you do detect…an exon 13 secondary mutation, sunitinib works well. If you have an exon 17 secondary mutation, sunitinib might not work that well, but you might want to test them a little sooner, keep a closer eye, and not go 3 months before the next scan, but probably do it in a 2-month timeframe.

Treatment Options in Gastrointestinal Stromal Tumors

When a patient presents to you with newly diagnosed GIST [gastrointestinal stromal tumor], there are certain things that are important to do in their work-up, and also in the approach to management. Patients with localized GIST sometimes present to surgeons initially, but at our center we’re good with multidisciplinary care. Even with localized GIST, it is important to involve the medical oncologist because many times we are able to help with neoadjuvant treatment to shrink the size of the tumors and make surgery easier, more amenable, more likely to get an R0 [margin negative] resection, and might even spare the patients a total gastrectomy and allow them to have a partial gastrectomy with more preservation of function.

With locally advanced GIST, of course, the multidisciplinary care is very important because you do want to treat them up front and get the surgeons involved at the right time when you’ve had maximum response with your oral therapies. Even then for patients with metastatic GIST, I frequently involve our surgeons in discussions, and we present them at the right time because patients with a limited metastasis, or if they’ve had a good response to imatinib, you always want to re-present to see if you can make those patients NED [no evidence of disease] with surgery so that they can continue on imatinib, or the TKI [tyrosine kinase inhibitor] that they’re responding on, for a longer time.

Where does molecular testing fit in? In my opinion, and at our center, prior to initiating TKI therapy, it is important to get molecular testing done. It is well recognized that 5% of patients will harbor a 

 resistant mutation that doesn’t respond to the frontline approved therapy of imatinib. It’s the 

 D842V mutation, and we have an approved drug for the treatment of that subtype. Especially in patients with locally advanced or limited metastases, treating with an effective drug up front can make a dramatic difference in the course of their disease and their ultimate survival outcome.

I do get molecular testing up front on all patients. It’s important to recognize the majority of patients, 85% of patients, will have a mutation in 

. You could do targeted testing only for 

 with specific exons that are common. In those who are negative for 

, that’s from 10% to 15% of patients, like this one that we just presented, then I do a more expanded NGS [next-generation sequencing] testing panel that will look at mutations such as 

 fusions that we have seen sometimes. The majority of those patients will have 

 mutations that form the majority of the so called wild-type GIST.

It is important, in my opinion, to get these genetic tests done. When we look at the 

 mutations, there are also implications for the type of mutation that is detected, which is predictive for therapy or response to therapy.

, the majority of mutations, 60% to 70%, are seen in exon 11, and do tend to be highly responsive to imatinib. Then the next most common in 

 are exon 9 mutations, which tend to require a higher dose of imatinib for treatment. For 

 mutations, the smaller minority are exon 12 mutations, which are similar to the exon 11 mutations in the juxtamembrane, and they do respond to imatinib well. It is the exon 18 mutations that are the most more common in 

, that tend to be resistant to imatinib and need an early evaluation and a look to see if a switch of therapy is required.

It’s also important to recognize that for frontline therapy we have only 1 approved drug, which is imatinib, and for 

 D842V mutations, avapritinib is approved in any line of therapy because that is still the most effective drug we have for that subtype. If you do end up detecting that mutation and if the patient’s already on imatinib and they’re not having a response, you can switch them, but if they haven’t started on imatinib, you can definitely start them on avapritinib in the front line.

As I told you, the majority of patients, 60% to 70%, will have a great response to imatinib. I tend to start them on imatinib because it takes sometimes a week to 2 weeks to get the mutation testing back. You go ahead and start them on imatinib, and once the mutation test comes back, if the result tells you that they are less likely to respond, and those would be the 

 D842V mutations, or even the wild type for 

 because sometimes they might not have as good of a response with imatinib. Those are the patients, even if they’ve started imatinib and they’re tolerating it well, I would do it early on—scan time point, maybe in 6 to 8 weeks—so that we can catch early progression and then switch to the next therapy that is appropriate.

Molecular Testing in Gastrointestinal Stromal Tumors

Hello. I’m Neeta Somaiah, a sarcoma medical oncologist at The University of Texas MD Anderson Cancer Center. One of my specialties is treating gastrointestinal stromal tumors. Today I’m going to be presenting a case, followed by a discussion. The case is of a 68-year-old man with a diagnosis of a gastrointestinal stromal tumor [GIST]. He initially presented with complaints of decreased appetite, vague abdominal discomfort, and a sensation of fullness that was preceding the 4 months of his presentation to clinic. His past medical history included diabetes that was medically controlled. He had had a colonoscopy at age 50 that was unremarkable, with no family history of cancer.

The physical examination showed abdominal pain on deep palpation, but was otherwise unremarkable. His clinical work-up previously included laboratory testing that showed a hemoglobin of 10.5 g/dL, his platelets were 105,000, and the other lab values were within normal limits.

He previously had had an endoscopy that showed a submucosal 6 cm mass with ulceration, and he had an endoscopic ultrasound following that showed a mass with irregular borders on the extraluminal surfaces with a marginal halo and hyperechogenic spots. The FNA [fine needle aspiration] biopsy showed a gastrointestinal stromal tumor with KIT on immunohistochemistry, and mitotic activity showed greater than 5 mitoses per 50 high-power field. This was sent out for targeted mutational testing and it was negative for 

The abdominal pelvic CT confirmed the 6.2 cm mass with indistinct margins in the body of the stomach. The MRI revealed 3 small hepatic lesions consistent with metastases. The diagnosis was stage IV GIST. His performance status was 0. He was started by his physician on imatinib, 400 mg orally once daily. It was continued for approximately 24 months until he complained of some increased abdominal pain and an additional 8-lb weight loss from his baseline.

His ECOG performance status had declined to approximately 1, the imatinib was discontinued, and he was initiated on sunitinib, 50 mg once daily for 4 weeks, with a 2-week drug-free interval. The treatment was initially well tolerated for around 3 cycles. Then he developed some nausea, emesis. The dose was reduced to 25 mg once daily, without resolution of his adverse events. Then the sunitinib was discontinued.

He was then switched to regorafenib 160 mg once daily for the first 21 days, with 3 weeks on for each 28-day cycle, so it was a 1-week drug-free interval. His repeat blood work showed the hemoglobin had dropped to 8.9 g/dL. His AST [aspartate aminotransferase] was elevated at 65 IU/L, and the regorafenib was discontinued. The patient was then started on ripretinib 150 mg orally once daily.

Is this the typical kind of patient I see? Yes. Quite often patents present to me in various lines of their treatment with metastatic GIST. If they present to me early, of course the mutational status is important for me to take into account. There are some additional tests I would do on their tissue if they were wild-type for 

In addition, as they progress through the different lines of standard therapy, I do sometimes dose sunitinib and regorafenib differently to improve tolerance. For those drugs sometimes I also start lower than the standard dosing that is approved because patients, especially elderly patients, or with poor performance status, have trouble tolerating the full dose. We tend to start low and then go high, and we’re able to get most people to tolerate it, at least in a lower dose, and a small fraction have to discontinue purely based on adverse effects.

This is the usual process of progression we see with patients with GIST. This patient had a wild type for 

, and this does tend to be a smaller percentage of the population of patients with GIST.

Neeta Somaiah, MD | Authors

Articles

Future Directions of Gastrointestinal Stomal Tumors

Understanding the INVICTUS Trial

Progressing Gastrointestinal Stromal Tumor

Treatment Options in Gastrointestinal Stromal Tumors

Molecular Testing in Gastrointestinal Stromal Tumors

Presentation of a Gastrointestinal Stromal Tumor Case