Greg Kennelty

On October 23, 2015, trabectedin (Yondelis) was approved by the FDA after promising results from the phase III ET743-SAR-3007 trial. Trabectedin is used to treat patients with advanced soft tissue sarcoma, including liposarcoma and leiomyosarcoma subtypes, and specifically those who have previously received chemotherapy that included an anthracycline.

In the open-label, multicenter, phase III SAR-3007 trial, trabectedin reduced the risk of disease progression by 45% versus dacarbazine in patients with advanced soft tissue sarcoma. There was also a slight survival trend with trabectedin, but the results were not significant.

The trial compared trabectedin with dacarbazine in 518 patients with liposarcoma and leiomyosarcoma who previously received an anthracycline-containing regimen followed by at least one additional line of chemotherapy. The protocol required patients to have an ECOG performance status of 0 or 1. The study was conducted at 85 sites in four different countries, but 94% of the patients were enrolled in the United States.

, Shreyaskumar R. Patel, MD, professor, Department of Sarcoma Medical Oncology, Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, and Jonathan C. Trent, MD, professor of Medicine, co-director, Musculoskeletal Center, Sarcoma Medical Research Program, Sylvester Comprehensive Cancer Center, University of Miami Miller School of Medicine, discussed the use of trabectedin and the future of the treatment paradigms in soft tissue sarcoma.

What have been some of the challenges toward evidence-based treatment in soft tissue sarcoma and how is that evolving?

One of the biggest challenges is that this is a rare group of tumors. Of the 1,650,000 or so new cancers diagnosed in the calendar year 2015, soft tissue sarcomas are just under 12,000 cases. As a group of diseases, it's rare. The added complexity is that the soft tissue sarcomas umbrella is comprised of 50 to 75 subtypes of sarcomas.

The three or four of the more common ones tend to be liposarcomas, leiomyosarcomas, and differentiated biomorphic sarcomas, but the list of entities that fall within this group of tumors that we call soft tissue sarcomas is very long. Consequently, what ends up happening is, to try to get level one evidence, you would need a large sample size in a very homogenous population. That's virtually impossible. Compare this to breast cancer, for example, where there are clearly more than 200,000 new cases in a given calendar year. The heterogeneity is a lot less than it is for soft tissue sarcomas. Consequently, it's not impractical.

As a matter of fact, it is standard practice to run between 2000 and 4000 patient clinical trials in the breast cancer population, in a uniform or homogenous population, to obtain the right answers that would constitute that level one evidence. That's virtually impossible in soft tissue sarcomas. You may run into several trials that may show a 10% difference in the primary endpoint in favor of an experimental arm versus the control arm, but the

-value may not be significant because the sample size is small. This is one of the major issues that we run into.

From a statistical standpoint, you really do want larger clinical trials and to get a large number of patients in a reasonable period of time. Of course, "reasonable" as defined by the sponsors and investigators, collectively. You then end up lumping a lot of patients together, and as I typically said, "you make a fruit salad out of the patient population." You mix apples, pears, peaches, and oranges, which makes the data that much less interpretable, if you will. To try to just test the apples, it will take a very, very long time, and that's the major limitation with soft tissue sarcomas.

Is there anything being done about it? I think over the last several years, the international community has come together. There are examples of trials that are run on both sides of the Atlantic, or even extending over to Asia, Australia, and beyond, to try to get behind a given trial that may be of interest to see if we can try to homogenize, if you will, the population as best as we can.

The management of sarcoma patients has always been challenging in terms of an evidence-based approach. There are a number of studies, but these clinical trials and other studies tend to be small. They tend to be limited in the focus on specific types of sarcomas. This is not surprising when there"s only 10,000 to 15,000 new patients per year diagnosed with sarcoma in the United States, and there are over 50 different types of sarcoma. It can be really challenging to have robust data and large numbers of patients.

That being said, there have been a number of randomized studies recently that have asked several specific questions. For instance, there was a recent study asking whether the use of gemcitabine/docetaxel was better than using an anthracycline such as doxorubicin in frontline therapy for patients with metastatic sarcoma. The result of that study was that doxorubicin was better tolerated and probably more effective than the combination of gemcitabine/docetaxel. Another study recently led to the approval of pazopanib in the treatment of patients with soft-tissue sarcoma, including liposarcoma and GI stromal tumors. This was based on a phase III study comparing pazopanib to placebo in patients with metastatic sarcoma.

A more recent study looked at the use of trabectedin in patients with liposarcoma or leiomyosarcoma. This subset of patients had metastatic disease, and the patients were randomized to either trabectedin or dacarbazine. The result of the study was superiority of the trabectedin arm.

You can see that there are a number of studies that are enrolling a larger number of patients, some of these in specific histologies such as the trabectedin study, which looked at liposarcomas and leiomyosarcoma.

What is the most effective method for a diagnosis and grading soft tissue sarcoma, and how important is molecular genetic testing?

The histopathologic diagnosis part of the disease is fairly well standardized. If patients are presenting with a lump or a mass, we will get a biopsy. The tissue is, and should be, analyzed by pathologists with expertise in musculoskeletal pathology. The characteristics of each one of these 50 to 100 different subtypes, depending on how many you're lumping under the umbrella of sarcomas, is reasonably well defined. Pathologists with expertise will not have difficulty in making the diagnosis.

I think the molecular genetics portion of this particular group of sarcomas is a different story. We will slowly, but steadily, be peeling away at some of the smaller subsets. Where there are characteristic translocations, for example, there may be fusion probings that can be identified to try to help from a diagnostic standpoint, assign a specific name to that particular entity.

Sarcomas have always been a challenging type of cancer to diagnose and grade. There are 10,000 to 15,000 new sarcoma patients per year in the United States compared with somewhere near 200,000 women diagnosed with breast cancer. This is a very rare disease. To make it more complicated, there are over 50 different types of sarcoma. In any individual type, there may only be several hundred or several thousand cases diagnosed per year in the United States. The rarity of this tumor makes it challenging for pathologists to make the diagnosis and assess the grade of the tumor.

At the University of Miami Sylvester Comprehensive Cancer Center, we have our sarcoma pathologists review all new cases in all new patients that come to our site. I think that the critical first step in understanding how to treat a patient with a sarcoma is making the right diagnosis. I generally recommend have the material reviewed at a sarcoma center by a sarcoma specialty pathologist.

In what population of patients would first-line chemotherapy be most appropriate?

I don't think these decisions are any different for soft tissue sarcomas than most other malignancies. Soft tissue sarcomas is more common in the fifth and sixth decade of life, so most of these patients tend to have a reasonable or a very decent performance status, even when they present with metastatic disease, organ function and performance status is acceptable enough for consideration of front-line systemic therapy. This is unlike some of the other epithelial tumors that may be far more widespread, where patients, either because of organ dysfunction, poor performance status, or just personal choice, would not be candidates for any line of systemic therapy. That proportion happens to be relatively small for the group of patients with soft tissue sarcomas.

In my practice, and in my experience, I think most of these patients with metastatic soft tissue sarcomas will end up going through multiple different lines of chemotherapy, or targeted therapies, or experimental trials, simply because they do tend to maintain and preserve their performance status and organ function for quite a long time.

Frontline chemotherapy is used in a variety of different situations in patients diagnosed with sarcoma. A patient is diagnosed with a primary tumor if the metastatic potential is high such as rhabdomyosarcoma, Ewing"s Sarcoma, desmoplastic small-round-cell tumor, osteosarcoma, these are patients that we initiate on chemotherapy as their frontline surgery, generally even before surgery or radiation therapy. In the metastatic setting, chemotherapy is the mainstay of treatment. These patients may get surgery or radiation at some point in the future, but in the setting of metastatic disease, the primary goal is to use systemic therapy to really control, shrink, hopefully even induce a complete remission for patients with metastatic sarcoma.

Who would be the right patient to receive trabectedin?

Trabectedin recently got approved for patients with metastatic or advanced leiomyosarcomas and liposarcomas. This was based on the way the clinical trial was designed. If you follow the prescribing information, then this would be appropriate for patients with leiomyosarcomas and liposarcomas who have metastatic disease, have advanced incurable disease, and who have failed at least one prior line of therapy that included anthracycline. This is the eligibility criteria for the clinical trial that led to the approval of this drug.

Certain subsets like a patient with myxoid liposarcoma are supposed to have exquisite sensitivity to this drug trabectedin, and if there were issues or contraindications to conventional anthracycline-based therapy, this [trabectedin] certainly can be moved up front in the line of therapy for that particular patient population.

Trabectedin is a new agent that we"re using to treat patients with metastatic or advanced sarcomas, specifically liposarcoma and leiomyosarcoma. Patients with liposarcoma and leiomyosarcoma generally present with primary disease, although many patients present with metastatic disease. Patients who do present with metastatic disease are often treated with an anthracycline combination therapy such as doxorubicin plus dacarbazine or doxorubicin plus ifosfamide. That"s standard frontline therapy for patients with metastatic sarcoma for most types. There are certainly caveats and certainly situations where that"s not the case, but that is the usual course of action.

After an anthracycline-based therapy, patients with metastatic sarcoma are treated with second-line therapy. In the setting of a diagnosis of liposarcoma or leiomyosarcoma, those options are gemcitabine/docetaxel for leiomyosarcoma or trabectedin, which has activity and is now FDA approved for leiomyosarcoma but also for liposarcoma, including dedifferentiated and myxoid liposarcomas.

What important developments have occurred toward improving treatments in leiomyosarcoma?

Leiomyosarcoma, one of the more common subsets of soft tissue sarcomas, has been a focus of attention of at least two large phase III randomized trials recently completed, presented, and to be published soon. One of them would be the Yondelis trabectedin approval for patients with leiomyosarcoma. The other one is a drug called eribulin. Eribulin, similar to trabectedin, was tested in a randomized phase III trial against dacarbazine and the data presented at ASCO last year showed improvement in overall survival. The drug is not yet approved for this indication, but certainly the hope would be that that becomes part of the therapeutic armamentarium for patients specifically with leiomyosarcomas.

Patients diagnosed with metastatic leiomyosarcoma have been difficult to treat over the years. Standard regimens include a combination of an anthracycline such as doxorubicin plus dacarbazine, occasionally doxorubicin plus ifosfamide. Second-line therapy that has been used in the United States includes gemcitabine/docetaxel, and now we have a new agent, trabectedin, that has been proven to be effective in second- and third-line therapy for patients with leiomyosarcoma. This is based on a phase III study where patients with metastatic leiomyosarcoma and liposarcoma were randomized to either trabectedin or a comparator arm of dacarbazine. Trabectedin was found to be superior.

What important developments have occurred toward improving treatment in liposarcoma?

I think both the phase III trials that we just mentioned included patients with liposarcomas and leiomyosarcomas. Yondelis is now approved for use in the United States, as it was for many other parts of the world. Eribulin data, like I said, was presented at ASCO and showed survival advantage, but is not yet approved by the FDA for this indication.

Articles

Shreyaskumar R. Patel, MD and Jonathan C. Trent, MD discussed the use of trabectedin and the future of the treatment paradigms in soft tissue sarcoma.

Evolving Paradigms in STS: Perspectives With Shreyaskumar R. Patel and Jonathan C. Trent, MD

The World Health Organization&rsquo;s (WHO) recently revised tumor classifications have made it clearer how to treat certain types of cancer, including diffuse gliomas and embryonal tumors, according to Arie Perry, MD.

, Perry, MD, chief of neuropathology at UCSF in San Francisco, California, discusses what oncologists should be aware of when diagnosing patients after these reclassifications, how it could change pathology, as well as a discussion on hereditary tumors.

Can you tell us a little about the changes in WHO classifications?

The biggest classification changes are occurring in the diffuse gliomas, which is a big category mostly in adults, and embryonal tumors, which is mainly occurring in pediatric patients.

What should oncologists be aware of with this change in classifications?

Oncologists should be aware that there are some new groups that are more accurately defined by molecular criteria and that if their pathologists are not doing the assays required to make those distinctions they probably need to have a chat with them or make sure that they send it out to laboratories where they are able to do that. It has prognostic significance and in some cases will probably change the management of the patient depending on the molecular results that are obtained.

So you mentioned that it could change pathology. What is being done with the pathology? Is there anything else that the new classifications can change for the action of the clinicians?

Yes, I think you know a lot of it is prognostic, so it may not necessarily change the therapeutic approach, but in some cases, if it&rsquo;s a dramatic change in terms of expected survival, for example, it may impact how they approach the management of the patient. In the embryonal tumors, the pediatric cases in particular, I think there&rsquo;s some really dramatic shifts in terms of either lesser aggressive therapies or more aggressive therapies depending on some of the findings.

Initially, in some of those patients, especially in the pediatric cases, there are actually germline mutations that we identify so it actually has implication for the entire family, in terms of hereditary tumor syndromes. So other family members may also be at increased risk of developing certain tumor types.

Can you talk a little more about the hereditary tumors?

In medulloblastoma, for example, there are several tumor predisposition syndromes that are associated with some of the youngest patients that get those. One is called Gorlin syndrome, and the patients may get some other things like skin cancers or basal cell carcinomas. It also may impact whether or not they&rsquo;re able to get radiation therapy, because actually it&rsquo;s contraindicated in many of those patients. If you give radiation, a lot of times they&rsquo;ll get skin tumors where the radiation is given.

There&rsquo;s another one called LiFraumeni syndrome that is due to p53 mutations, and other family members may be predisposed to get colon cancer, or breast cancer, or sarcomas, which are malignancies in the soft tissues in the bodies.

What should we be looking for in the future with kind of how tumors are classified, and what it means?

I think more and more we&rsquo;re adding genetics into our classification scheme and probably more and more in the future we&rsquo;ll also be using genomics and rather than looking at one gene at a time, or one biomarker at a time, we&rsquo;ll actually be looking at the entire genome or at least cancer-associated genes within the genome,. We&rsquo;re already starting to do that, especially within the pediatric cases, but a subset of the adult cases as well, and so sometimes we&rsquo;ll find mutations or genetic alterations that aren&rsquo;t necessarily reported in that tumor type, but for which there are targeted therapies, so it may shift the way that the oncologists then decided to treat that patient.

I think the takeaway is that things are changing in a good way in terms of it becoming a little more complicated and more technical. But it used to be somewhat subjective and there were differences between one pathologist diagnosis and another, and I think the more we define things objectively with molecular markers, the less that&rsquo;s going to occur and the more confidence the oncologists will have that the diagnosis is accurate, and more importantly perhaps will guide the way they treat individual patients, so it won&rsquo;t necessarily be that everyone with the identical diagnosis will be treated the same way because there may be other differences in the molecular makeup of that tumor.

What does the multidisciplinary approach look like at the tumor board?

It&rsquo;s the same way its been for many years, although you know everyone&rsquo;s area is becoming more complicated. I think it&rsquo;s even more important that we get together once a week at least to go over patients and the intricacies of the pathology, the diagnosis, changes in radiology, what protocols may have just opened up, what&rsquo;s available for certain patients, what molecular studies have shown us that may be expected or unexpected in that particular patient, and then what&rsquo;s the best approach for this patient going forward.

Arie Perry, MD, chief of neuropathology at UCSF in San Francisco, California, discusses what oncologists should be aware of when diagnosing patients after these reclassifications, how it could change pathology, as well as a discussion on hereditary tumors.

The Impact of WHO's Tumor Reclassifications on Brain Cancer Care

-mutated colorectal cancer may help formulate new patient-specific treatments, according to Maria Ignez Braghiroli, MD.

, Braghiroli, medical oncologist, Instituto do C&acirc;ncer do Estado de S&atilde;o Paulo, discusses a study that parsed out patients with solely

-mutated colorectal cancer, what the current treatment paradigm is for these patients, and how the mutation status of a patient with colorectal cancer may affect their treatment.

Can you tell us about clinical characteristics of patients with metastatic colorectal cancer harboring

We've been trying to molecularly characterize the type of tumors, and possibly try to individualize treatment. What we did is we selected all the colorectal cancer patients that were found to have

mutations from 2008 to 2015 that were treated at Memorial Sloan Kettering. We found a total of 87 patients, and we had to exclude 3 patients because they had concurrent

mutations and we wanted the patients with

mutations only. Then we compared this population in terms of clinical characteristics, molecular characteristics, and outcomes to treatment in the population with

mutations. How would that have affected the results?

We don't really know, and that's why we excluded only 3 that had both the

mutations. That's why we didn't include them in the

What would you say the current treatment paradigm is for

Basically, we know that anti-EGFR medications don't work for

-mutant patients. These are the smaller proportion of patients with colorectal cancer, around 3% to 7%. That's why we don't really know what happens to this specific subtype of

mutation patients, and that is why we were looking into it.

How much would you say mutation status affects the individual treatment of patients with colorectal cancer?

mutated patients, we know that they do not respond to the anti-EGFR mutations. In general, we know that maybe

-mutated patients have a much worse prognosis, and the studies currently focused on

inhibitors or MEK inhibitors, or a combination of

, MEK, and anti-EGFR inhibitors in these patients. That is actually being performed at Memorial Sloan Kettering also. A small subtype of patients actually have an amplification of HER2, for example, and they might benefit from anti-HER2 treatment.

There is also a sub-population that has its microsatellite stability, and maybe this population benefits from immunotherapy. So we are tending to find a small, specific subtypes of tumors, and then individualize their treatment in this way.

Are there any ongoing or planned studies in this space that you are particularly excited about?

There are many studies in immunotherapy. I think this is very interesting, and so far for the sporadic type of colorectal cancer, we seem to not have very good results, so I think studies focusing on that will be very interesting.

What do you hope to see in the next 5 to 10 years in the treatment of colorectal cancer?

Immunotherapy is one thing. I think we might understand left or right sided, and if this is really important for patients and if that has any implication for treatment. Hopefully, we'll have newer medications with more consistent improvement in survival, and maybe earlier detection via colonoscopy. It's not an easy procedure to perform for the patients, and there have been ongoing studies on fecal identification of tumor cells. That, or maybe blood identification of tumor cells to try to find them earlier.

What role does sidedness play in colorectal cancer?

There has been some evidence that patients with right-sided tumors may have a worse prognosis than patients with left-sided tumors. In our population with

-mutated patients, the patients did worse than all

wild-type tumors, but interestingly, they had more left-sided tumors than right-sided tumors. This is still a matter of research.

Maria Ignez Braghiroli, MD, discusses a study that parsed out patients with solely NRAS-mutated colorectal cancer, what the current treatment paradigm is for these patients, and how the mutation status of a patient with colorectal cancer may affect their treatment.

Expert Discusses Treating Patients With NRAS-Mutated Colorectal Cancer

The GeparSepto trial shows nab-paclitaxel (Abraxane) proves beneficial over paclitaxel in certain subsets of breast cancer, according to Sibylle Loibl, MD, PhD.

, Loibl, co-chair of the German Breast Group (GBG) and associate professor at the University of Frankfurt, discusses the most recent subanalysis of the GeparSepto trial, how the drugs match up in terms of toxicities, and a separate set of data looking at patients with breast cancer who are also pregnant.

Can you tell us a little about the subanalysis of the GeparSepto trial?

The GeparSepto study was investigating if the exchange of nab-paclitaxel instead of paclitaxel followed in both arms by epirubicin (Ellence) would lead to an increased pathological complete remission (PCR) rate. This could be demonstrated overall, but we were curious to see how that would play out in the subgroups of luminal patients, triple negative breast cancer patients, and HER2-positive with or without hormone receptor (HR) expression. The patients with HER2-positive disease received, in an addition to the chemotherapy, pertuzumab (Perjeta) plus trastuzumab (Herceptin) from the very beginning for the whole 24 weeks and continued afterward with trastuzumab for up to 1 year.

We have seen that in the HER2-positive cohort, the PCR rate was much higher. We achieved a PCR rate of almost 75% in the HR-negative HER2-positive cohort. There was a difference in favor of nab-paclitaxel, but it was not statistically significant. In that group, the HER2 treatment is probably more important than the chemotherapy schedule. What we also found, and this has already ben seen in previous studies, is that the definition of the PCR is more important in the HER2-positive cohort than the triple negative cohort.

The second results were about the toxicity, because there have not been so many data in the neoadjuvant setting using the 2 antibodies. In fact, this is the largest cohort using the 2 antibodies in the neoadjuvant setting and using 24 chemotherapy regimen. In TRYPHAENA, the chemotherapy was 18 weeks. We found the PCR rates were comparable to that of the TRYPHAENA trial when pertuzumab was added to trastuzumab, and the toxicity was also comparable to that. Especially the rate of diarrhea is to be noted, so up to 10% of the patients had greater than grade 2 diarrhea, which seems to be noteworthy.

Are there any next steps or other analyses of this trial that are important?

We're currently conducting the rate of peripheral sensory neuropathy, and how that has evolved after patients have completed their treatment. We have seen a higher rate with peripheral sensory neuropathy with paclitaxel compared to nab-paclitaxel, but we can also see that the neuropathy is decreasing around 1 year after the treatment, and the rate then of grades 2 and higher is equivalent between paclitaxel and nab-paclitaxel. Grade 3 and 4 is a little more tricky because the neuropathy stays a little longer.

At the moment the neuropathy fades away. There have been several studies investigating ameliorating treatment and supportive treatment, but so far neither has really been proven to be helpful to prevent or treat it.

What do you really want oncologists to be able to take away from this analysis?

What we have seen in the GeparSepto study is the greatest benefit of using nab-paclitaxel instead of paclitaxel was actually in the triple negative cohort, where we could almost double the PCR rate. This is a very important message because these are the patients currently treated with chemotherapy, and therefor you need to use the best chemotherapy that you have. it is less critical in the luminal and HER2-positive type. So we would like oncologists to take this message in the triple negative population and use nab instead of paclitaxel, and monitor the periphery sensory neuropathy carefully.

Do you think there will be a point where we combine the nab-paclitaxel with other agents?

Probably the most interesting combination so far are the checkpoint inhibitors, which there are trials investigating. There are also data combining nab-paclitaxel with carboplatin in the firstline setting and the metastatic setting, though there's a trial looking specifically into this combination. These are the two main combinations currently being investigated in triple negative breast cancer.

Can you tell us a little about your collection of data from pregnant patients with breast cancer?

We have, since 2003, a registry ongoing collecting patients diagnosed with breast cancer during pregnancy. In parallel, we are also collecting data from patients who are very young with breast cancer, meaning diagnosed below the age of 40, as a control cohort. We collected tissue as well from our pregnant breast cancer patients, and we investigated if there is a different mutation pattern in the pregnant breast cancer patients compared to data which we retrieved from already collected data set.

In essence, we found the main driver is probably the different cohorts. The pregnant breast cancer cohort was younger, and there were less HR-positive patients in this cohort compared to the TCHA dataset after matching the numbers, although this is the largest pregnant breast cancer cohort investigating somatic mutations in breast cancer. We also found that we have higher amounts of TP53 mutations in the pregnant and less PI3 kinase mutations, which could possibly explain the slight imbalances in the baseline characteristics.

What kind of clinical implications can this data have?

We cannot derive any clinical conclusions from that. There was the rumor that patients diagnosed with breast cancer during pregnancy have worse outcomes due to the underlying biology. So far in the large dataset from our cohort of almost 500 patients and 800 controls, we could not demonstrate that the pregnant patients had a worse outcome. The worse outcome some other investigations have shown is mainly driven by the non-standard treatment of those patients. We wanted to see if we can see some biological differences by investigating the mutations in this cohort.

Sibylle Loibl, MD, PhD, discusses the most recent subanalysis of the GeparSepto trial, how the drugs match up in terms of toxicities, and a separate set of data looking at patients with breast cancer who are also pregnant.

Greg Kennelty | Authors

Articles

Evolving Paradigms in STS: Perspectives With Shreyaskumar R. Patel and Jonathan C. Trent, MD

The Impact of WHO's Tumor Reclassifications on Brain Cancer Care

Expert Discusses Treating Patients With NRAS-Mutated Colorectal Cancer

Nab-Paclitaxel Beneficial Over Paclitaxel In Breast Cancer Subsets