Login | Register

Research Findings in Carcinosarcoma: Interview with Dr. Whitfield Growden


The following transcript is from an interview between Dr. Whitfield Growdon (MDC), a leading GYN Oncological Surgeon from Mass General Cancer Center in Boston and Diane Redington (DR), a woman with a Gynecological Carcinosarcoma (GCS) who is seeking to motivate researchers to advance the science and uncover knowledge about carcinosarcoma and find a cure for this cancer. On January 5, 2017, Diane conducted a series of interviews with Dr. Growdon on the topic of carcinosarcoma. This interview discusses diagnosis and treatment of MMMT.

Dr-Whitfield-GrowdonDr. Whitfield Growdon was born and raised in Boston. He graduated from Williams College with a degree in Art History. After years of appreciating fine arts in the Berkshire Mountains, he attended the University of Massachusetts Medical School. He returned to Boston where he was a resident at the integrated Brigham & Women’s/MGH Combined OBGYN Residency and also completed a gynecologic oncology fellowship at the Massachusetts General Hospital. His interests include minimally invasive and robotic surgery, correlative investigation into rare gynecologic tumors and innovative clinical trial development.

Research Findings in Carcinosarcoma

Dr. Growdon, thank you so much for sharing your knowledge and expertise with the women suffering from this horrid cancer. Can you tell us current scientific findings on carcinosarcoma?

What we did was, we looked across all of the GYN carcinosarcomas because we had a sneaking suspicion that maybe they’re all related, and when they happen to pop up, is just random entropy [lack of predictability]. It doesn’t necessarily reflect how we should be treating it.

Or maybe the molecular drivers that make carcinosarcomas happen are the same in the fallopian tube, in the ovary, in the uterus and in the vagina.

And so we were able to use the vast tumor bank of Mass General to obtain tissue. Then we actually interrogated [examined] the tissue for specific gain-of-function oncogenic mutations. We queried it for about 150 or so mutations and basically distilled each individual tumor down to its DNA and RNA and were able to genotype each one. What we found was fascinating.

  • Fallopian tube carcinosarcomas are very, very rare, but they tend to look just like ovarian carcinosarcomas.
  • Vaginal – we only had 2 [vaginal tumor specimens] out of our entire cohort of approximately 70 to 80. So it was hard to draw firm conclusions. We did not see any clear pattern of gain-of-function mutation.
  • It was the uterus where we found over 50% gain-of-function mutations in uterine carcinosarcomas. In uterine cancer, there are defined subsets of specific types of tumors that have specific mutations that we believe make these tumors very susceptible to targeted therapies as well as immunotherapies. Carcinosarcoma is just at the tip of that.
  • Ovarian and fallopian tube carcinosarcomas did not have any meaningful mutations in specific subsets, which really mirrors regular serous ovarian cancer. Serous ovarian cancer isn’t marked by specific subsets of tumors that have treatable targetable mutations. It’s marked by random genetic entropy – random, what we call genomic instability. Where the rule is, “There is no rule in what the molecular changes are.”

Carcinosarcomas have historically been excluded from clinical trials

epithelial-ovarian-cancerOne of the reasons why is that in 2009 we changed carcinosarcomas from being a “sarcoma” classification to an “epithelial carcinoma” classification. So the cancer genome atlas, all of the major clinical trials in carcinoma of the uterus, excluded carcinosarcoma. They were excluded for years. It has only been in the last 5-7 years that carcinosarcoma has been included with epithelial cancers. We are starting to think about them very differently. This is largely because of work done in the early 2000s that basically looked at molecular underpinnings of carcinoma as well as carcinosarcomas and found that what they think happens with carcinosarcomas – and this may be true of ovarian, we don’t know, but it is probably true of endometrium – is that they started as a high-grade carcinoma like a serous carcinoma and as they grow, they de-differentiate or morph into something completely different at the same time. What it looks like under a microscope is a tumor that has both sarcoma elements and carcinoma.

Carcinoma or Sarcoma?

They used to call these back in the 80’s, “collision tumors.” As if it were 2 different cancers colliding together. A lot of researchers, without knowing the molecular underpinnings, assumed that the sarcoma was driving it, when in fact it is probably the carcinoma that is driving it. That was really on the epidemiological level or on a clinical level seen, because when these cancers metastasized to different parts of the body, it was usually the carcinoma that metastasized. We thought more than likely to be “the egg and not the chicken.” We don’t have definitive proof, but we believe now, based on our molecular studies of lineage, that likely these carcinosarcomas – the sarcomatous components – are derived from the epithelial components. Even in our study where we biopsied the sarcoma and biopsied the carcinoma, they found excellent homology [the same or similarity] at the genomic level between those two things.

Why not go after both?

Response to therapy

The good news here is that it looks like the molecular alterations that affect the carcinoma also affect the sarcoma, and it is just the way they have differentiated that makes them different. Which means that if you can find a therapy that addresses the carcinoma, like a targeted therapy or an immunotherapy, it is likely that you are going to also hit the sarcoma as well.

Which is where carcinosarcomas are different from pure sarcomas, like rhabdosarcomas, like liposarcomas. There is a huge family of sarcomas that get treated with very different chemotherapy than we use with carcinosarcomas.

And so this is something that’s definitely evolving. This is something that I think there have been dedicated clinical trials — not for ovarian carcinosarcomas, but for uterine carcinosarcomas. It was recognized that this was a separate type of cancer early, like in the 90’s.

Ovarian carcinosarcoma clinical trials

I think that for ovarian carcinosarcomas, as we talked about briefly, a lot of times most of our clinical trials, unfortunately, have excluded ovarian carcinosarcomas because they wanted to create pure populations for clinical trials. There just weren’t enough, basically, incident cases to fill a trial and possibly have an impact. So they got excluded, and a lot of our ovarian cancer treatment paradigms for carcinosarcoma are based on serous epithelial ovarian cancer paradigms. Whether or not that is relevant, we don’t know. We are just not sure.

Whereas in uterus, there definitely is. Dedicated trials. There are some ongoing clinical trials right now.

The big trial that we are waiting for compares two strategies of conventional chemotherapy. One is using a medicine called Taxol along with a medicine called ifosfamide. That was established as our standard of care with the best response rates and the best overall survival when compared to other regimens. We are now comparing that to carboplatin and Taxol. So these two different regimens that both have Taxol are being compared. We currently use carboplatinum (carboplatin) and Taxol as our front line. A lot of times ifosfamide has fallen out of favor. It can be onerous to give. It is given over several days, has a toxicity profile that is more than what we see with every 3-week carboplatin and Taxol. We think that based on smaller studies that likely the response rates are the same if not maybe a little bit better. So we are confident that that trial will likely show equivalency or at least non-inferiority with much less toxicity.

Is that for all carcinosarcoma?

That is for carcinosarcoma of the uterus.

Carcinosarcoma of the ovary does not have, as of yet, dedicated clinical trials where you just have women that just have carcinosarcoma of the ovary. Some trials lump them in, but most exclude them.

Important Differences

They are 2 different entities, so it is important to differentiate between the diagnosis of ovarian carcinosarcoma versus uterine carcinosarcoma.

I think that is a very important difference. It is a differentiation that sometimes is very hard to make.

Would it be based on a tissue sample?

It would be based on a tissue sample, and you look and see where things arose. We know that on a molecular basis, there probably are differences. If you were ever wondering, “did this come from the ovary or did this come from the uterus?,” it would be nice to have a test you can do. Some sort of molecular analysis. But the fact of the matter is there is likely not going to be a specific test. We are never going to be able to differentiate them perfectly.

Imaging Studies

If you are diagnosed with carcinosarcoma and you are not a surgical candidate, how would you be able to differentiate between an ovarian and a uterine carcinosarcoma?

That’s a great question.
A lot of times we would want to do advanced imaging studies.

Based on a pattern of metastasis?

No, probably not based on a pattern of metastasis, although that might help. A lot of times you would want to image the pelvis very carefully. Our practice here, whenever I take care of a new person that has carcinosarcoma suspected either of the uterus or the ovary,

  • I consider getting a PET CT, generally up front. Generally with uterine carcinosarcoma, we know that the chances of having metastasis early is higher.
  • I do the PET CT to understand what’s going on in the neck, the chest, the abdomen and the pelvis.
  • The PET CT will administer a sugar molecule that is attached to a phosphor that would glow brightly in the PET scanner. Cancer cells are much more likely to take up sugar, so that in theory, things that glow bright will look positive possibly for cancer.
  • It helps me actually map out what type of surgery I am going to do. A lot of times you can see abnormalities on the ovaries, and you see abnormalities within the uterus.
  • For deep pelvic imaging, I will generally get an MRI. Sometimes I do both because I want to understand where this growth appears to be coming from.

We still see metastasis to the uterus and even synchronous primaries sites [two different types of cancer occurring at the same time]. We see abnormalities on the ovary simultaneously with the uterus.

Is it synchronous metastasis?

Right, and you cannot tell if this started in the ovary and go to the uterus or did it start in the uterus and go to the ovary. Which is why we use broad terms like Gynecologic Carcinosarcomas. I think an important differentiation that you are getting at is treatment might change based on whether or not you have an ovarian carcinosarcoma or a uterine carcinosarcoma.

Diagnosis and Treatment of Carcinosarcoma

Could you talk about that? If you are diagnosed with an ovarian carcinosarcoma, could you take us through the stages of what you would do? When is it appropriate to have surgery versus when surgery is not useful? Would you talk about the different treatment modalities?

Ovarian Carcinosarcoma

If we suspect that this is an ovarian carcinosarcoma, our treatment paradigm essentially mirrors that of epithelial ovarian cancer. Generally, we do:

  • A good exam
  • Good lab value testing – things like CA125, Carcinoembryonic Antigen (CEA). We sometimes get a CA19-9 [Carbohydrate Antigen 19-9]. We do that for all of our patients with any type of epithelial or suspected epithelial ovarian cancer.
  • Then we will get imaging because what we want to know is, are we able to do an upfront surgery or an interval surgery that will allow us to remove all the cancer cells?
  • Sometimes ovarian carcinosarcomas affect tissues that we cannot remove, and it would not be safe to remove. So I would not recommend upfront surgery in that setting. I would want to give chemotherapy first.


  • The standard treatment for an ovarian carcinosarcoma would be giving conventional carboplatin and paclitaxol (Taxol) therapy.

Intraperitoneal Chemotherapy

  • A lot of people wonder about intraperitoneal chemotherapy. This is something that we offer to women who have a Stage 3 ovarian cancer that got surgery first and had almost all of their cancer cells removed during surgery. I will commonly put in an intraperitoneal port, which is a port that sits on the rib cage either on the right or the left. With regards to intraperitoneal chemotherapy, this is a purely ovarian cancer modality that we use. And it is invariably used when we do an upfront surgery, remove the vast majority of cancer cells that we can see with our eyes, and then we put a port in on the right or the left, and then we administer chemotherapy both in an IV as well as intraperitoneal in order to apply chemotherapy directly to the surfaces that are most likely to have residual microscopic cells. In several trials, this was associated with progression-free and overall survival benefits – by giving intraperitoneal as opposed to purely IV. This is controversial, because all of those trials excluded women with carcinosarcoma.
  • Our practice here is that we look at that and say, on a genomic level, ovarian carcinosarcomas appear to look like epithelial ovarian cancers, so we offer women intraperitoneal chemotherapy. I have personally taken care of women who have done very well with intraperitoneal chemotherapy, and they have had as excellent outcomes as what we have seen with non-carcinosarcoma epithelial ovarian cancer. Again, because we think it is probably a variant of epithelial serous ovarian cancer, which is why we see good responses.

This is would only be administered following a surgical procedure?

With intraperitoneal, that is correct.


Is surgery indicated for Stage 1, stage 2?

A lot of times for Stage 1,2, and 3, we will offer a surgical procedure upfront with ovarian cancer, assuming their imaging looks like we can remove all the cancer cells. With Stage 4 disease, when we see evidence of either metastasis within the liver in places where you cannot remove or…

Within the liver as opposed to a coating on the liver?

Chemo followed by Surgery

Correct. Or we see chest disease, meaning basically lung metastasis, or brain metastasis or something that wouldn’t be addressed with a surgical procedure in the abdomen, many doctors will go ahead and give chemotherapy up front, see a response in the chest or wherever else it is extra-abdominally and then pursue a surgical procedure in the abdomen. Once we feel like these cancer cells here have responded and are shrinking and we have good therapy for that, then we would want to actually “cyto-reduce,” is what we say. Basically lower the burden of cancer cells in any given person so that the chemotherapy can be more effective.

What about a para-aortic lymph node metastasis? Would that eliminate surgery?

No, it would not. That would be true of uterine carcinosarcoma as well as ovarian carcinosarcoma.

Uterine Carcinosarcoma

A normal paradigm for me when I take care of women who have uterine carcinosarcoma is that:

  • I will get a PET CT, which will tell me about para-aortic lymph nodes, the pelvic lymph nodes and
  • Then what I will do is, I’ll do a comprehensive laparoscopic staging procedure, which is where we remove the uterus, the ovaries, the fallopian tubes, the pelvic and para-aortic lymph nodes and test them.
  • We know that the rate of lymph node metastasis is higher, sometimes upwards of 40% of the time with these high-grade carcinomas. And this helps us direct our future therapy.
  • Even with Stage 1 uterine carcinosarcoma, I recommend treatment with chemotherapy, which is not true with other types of uterine cancer. We treat uterine carcinosarcomas differently. We tend to give them more chemotherapy up front.
  • In the setting of lymph node metastasis, we also apply radiation therapy. This is targeted radiation therapy.

That is another major difference in paradigm between ovarian carcinosarcoma and uterine carcinosarcoma. With uterine cancer, we have a much greater experience using radiation therapy. We know that it can be a very important adjuvant that can reduce recurrence risk.
So surgery and possibly intraperitoneal chemotherapy.

For ovarian, correct.

For ovarian — and then ongoing chemo?

We usually do 6 cycles.

Only 6?

In the upfront setting, we generally will give 6-7 cycles of chemotherapy for ovarian carcinosarcoma or uterine carcinosarcoma. Then we will follow that up with some sort of an assessment, like an imaging study to understand if there is anything left over. Is there any residual [cancer] that we are concerned about? If we don’t see anything, we like to stop therapy. The reason for that is we don’t want to tirelessly do more and more chemo. There are some older trials that suggested doing more chemo if there is no target will just exhaust people and decreases a woman’s functional status. So that such that if this were to come back, they may not be able to tolerate as much chemo. So a lot of times we think about stopping. If we see something, then we talk about additional therapies. We will talk about changing things and seeing whether or not we should be giving additional therapies.

Why stop chemo if the cancer usually comes back?

But the thing is that it always comes back. It always returns. So that window you have where you stopped chemotherapy now gives those microscopic cells the opportunity to grow. It is rare for it not to come back. So I think, why not just have some ongoing maintenance?

So, a lot of maintenance strategies have been looked at. You are absolutely right. The real Holy Grail of oncology would be to give a medicine that gets rid of everything and then give a less toxic dose that people can live on. Like Tamoxifen for breast cancer. That was a blockbuster, because it helped women. They were able to take something that meaningfully reduced their recurrence risk. We haven’t found that for any type of ovarian cancer. We haven’t found that for any type of uterine cancer. And so carcinosarcoma, which are more uncommon versions of both of those things, are probably even further off from having something like that.

Right now, the therapies that we do have, women cannot be sustained on. Even something like single agent Taxol was looked at in ovarian cancer. And what they found was that if you gave women a years worth of single agent Taxol after finishing upfront therapy, you did delay any recurrence, but you never stopped it. And when women developed that recurrence, it was unclear if they had enough reserve to do more necessary chemotherapy. And so you are bringing up the most important dilemma that we have.

The treatment failure and the recurrence rate is so high with these cancers, why not just start off with all guns blazing? With a combination of chemotherapy, immunotherapy — give it all you got. What do you have to lose? Most of these women are going to die.

You bring up a very good point. Right now, the paradigm for ovarian cancer, meaning that we sort of have this clinical divide between the ovarian carcinosarcoma and the uterine carcinosarcoma:

  • And 50% of women with uterine carcinosarcoma have Stage 1 disease, meaning it is confined to the uterus. But even though 50% have Stage 1 [in uterine carcinosarcoma], in regular endometrial cancer, 85% are Stage 1.
  • 85% with uterine carcinoma are diagnosed at Stage 1. So the majority of endometrial cancer as a cohort – the 64,000 women diagnosed with endometrial cancer – the majority of them, 85% are Stage 1. [Remaining 15% will have advanced disease]
  • But in carcinosarcoma, it changes. 50% of women with uterine carcinosarcoma have Stage 1 disease, meaning that it is confined to the uterus. [Remaining 50% are diagnosed with advanced disease.] But even those Stage 1s have a much higher recurrence rate. With stage 1 endometrial carcinosarcoma, we give chemotherapy. And we don’t do that for other types of endometrial cancers. So we do get more aggressive. We are also more likely to give adjuvant [aiding, supportive] radiation therapy. Those are the two modalities that have been most tested. We do tend to throw the kitchen sink at carcinosarcomas.

Unfortunately, there are clinical trails that have looked at this and have sort of lumped carcinosarcoma in with a group of higher-grade endometrial cancers. What they found is that they haven’t been able to meaningfully change the recurrence rate. Meaning the kitchen sink we have is not getting the job done. It’s tiring women out. Women feel like they are getting hit by the kitchen sink, but it doesn’t appear to be moving the bar where we need it.

Watch the two videos of the transcript