Randall Reviews Rare Incidence of and Challenges With Spinal Sarcomas

R. Lor Randall, MD, discusses the data from the population-based registry and provided more insight on prognostic factors and survival outcomes for patients with spinal sarcomas.

Sarcoma of mobile vertebral column continues to be a rare finding in patients, but is one with poor survival outcomes and varying prognostic factors, explained R. Lor Randall, MD.

Because primary sarcomas that originate in the spine are so rare, there are inconsistent data on this patient population, which is also very heterogenous.

In an analysis of a population-based registry, investigators utilized information from 1975 to 2017 in the Surveillance, Epidemiology, and End Result (SEER) database to gather incidence and survival information in 712 US patients. For prognostic factors that impacted survival, Kaplan-Meier and Cox Regression were used as measurements.

Data showed that the incidence of spinal sarcoma was 0.019 per 100,000 persons in 2017, which has not changed significantly since the year 2000 (P >.05); the 5-year disease-specific survival rate was 57%.

When stratified by histology, chondrosarcoma had the best 5-year overall survival (OS) rate at 69% compared with osteosarcoma at 39%. Age, grade, and stage were independent predictors of survival in the entire cohort. More specifically, age was a predictor of survival in spinal osteosarcoma, size for Ewing sarcoma, and grade for chondrosarcoma. 

In an interview with OncLive, Randall, David Linn Endowed Chair for Orthopedic Surgery, as well as professor and chair of the Department of Orthopedic Surgery at University of California Davis Comprehensive Cancer Center, discussed the data from the population-based registry and provided more insight on prognostic factors and survival outcomes for patients with spinal sarcomas.

OncLive: Please discuss the analysis you conducted on the incidence of spinal sarcomas.

Randall: We work very much hand in hand with our spines team, so we do combined efforts. If there's a primary sarcoma involving the spine, we usually do it as a joint surgical procedure. The technology has gotten so advanced that if you're not doing spine all the time, you just can't keep up with it. It has evolved so, so much.

Having said all of that, we have just been looking at this SEER database, just to see if there's anything new that's popping up. We have been mining it; there were some papers that came out looking at SEER data. We thought it had been a while; we hadn't seen anything recently. We decided to look at the SEER data and see up to 2017 whether or not anything new was being discovered in mobile vertebral column or spine.

What we found that was interesting was that there has not been an increased incidence; it's pretty plateaued, which is not really a surprising finding. It's still very, very uncommon. The most common types are Ewing sarcoma, then chondrosarcoma, and then osteosarcoma. Patients with chondrosarcoma tend to do the best and that's most likely due to the fact that many of the chondrosarcomas are of lower grade. Patients with Ewing sarcoma tend to do better than those with osteosarcoma, and that's probably because of the radiosensitivity of Ewing sarcoma. You can imagine if someone presents with a Ewing sarcoma of the spine, because the spinal cord or the nerve roots are right there, it's hard to get a margin on it. You need radiation to help control the local disease, and Ewing sarcoma is radiosensitive, whereas osteosarcoma is radio-refractory.

What remaining questions do you have regarding this research?

We weren't able to find, unfortunately, what we really wanted to look at: health care disparities. What we really want is to look back at insurance status and socioeconomic status as markers for social prosperity, so to speak—to see how patients do. We have some other research projects that we're looking at [with sarcoma in] the hand and the foot, where there is more of an incidence and more data, where we will be able to start drilling down on whether or not socioeconomic variables are prognosticating of outcome.

However, we weren't able to do that with this study. There just wasn't enough data.

Why is there a higher incidence with sarcoma in the hand in the foot?

Well, they are still relatively rare. These are all very uncommon, and sarcomas tend to arise in larger tissues that have more anatomic area. If it's a bone sarcoma around the distal femur, proximal tibial, or proximal humerus, those are sort of the largest bones, and they have the biggest growth weights in there for the most metabolic activity. These other areas don't tend to be too common because there's smaller growth potential etc. Because they're still in the extremity, there seems to be a bit more information collected on extremity sarcomas than we're able to collect on the spine circumference.

What is the optimal approach for managing a spinal sarcoma?

It for sure requires a multidisciplinary approach. The technology around spine reconstruction requires spine expertise. There are so many orthopedic oncologists who have historically done a lot of spine surgery, but unless they're doing a lot of high-volume surgery, the message is that it really should be multidisciplinary. Work with your spine team.

The other message is that the incidence is remaining constant, but we are finding that patients with chondrosarcoma do the best, following by Ewing sarcoma, followed by osteosarcoma. A lot of the tumors arise present pretty late, and that's because a lot of patients without tumors have back pain. Therefore, for patients with tumors who present with back pain there's not a degree of vigilance—and in no way am I suggesting that everyone with back pain get worked up for a tumor. However, [I'm talking about] the atypical pain: the pain at night, the pain not associated with activity, the pain that is not getting better.

Obviously, starting with simple plain radiographs is important. However, if there's something on that radiograph or one should work it up with an MRI pretty vigilantly. The other thing to point out is if it happens in younger people, which is what happens with osteosarcoma and Ewing sarcoma; young people tend not to have a lot of back pain. If they are younger, and they're not getting better after some conventional modalities, you want to perhaps be a little bit more concerned about something more aggressive going on.

Is neoadjuvant therapy an approach you're doing for these patients?

That is a great question. Neoadjuvant treatment is certainly used for Ewing sarcoma and osteosarcoma. In fact, we argue pleasantly about Ewing sarcoma—whether it should be treated primarily with radiation or surgery in terms of the local control.

With both Ewing sarcoma and osteosarcoma, patients get neoadjuvant chemotherapy, and then when it comes to local control, for osteosarcoma, it's exclusively surgery. For Ewing sarcoma, sometimes it's surgery and or radiation. In America, we tend to do surgery or radiation; in Europe, they're much more inclined to use combined modality.

We argue about that quite a bit constructively, because the upside is you might be able to affect local control. The downside is now you're introducing 2 toxic therapies: surgery and radiation. Another important consideration with this is a significant portion proportion of patients will present with neurologic symptoms.

I mentioned earlier about back pain, but sometimes they will present with motor weakness, cauda equina syndrome, or dermatomal sensory abnormalities that are unusual. In those cases, sometimes [the patients are] not worked up vigilantly enough.

These patients go on to develop chord compression of significance that becomes an emergency. Obviously, local control of a sarcoma is of paramount importance, but paraplegia and quadriplegia are pretty darn important, too. Sometimes, when [the sarcoma is] very advanced, the surgeon is forced to go in and decompress the spine. In that moment, they're more concerned about decompressing the spine than they are in getting adequate local control of the tumor, understandably, so we then have a compromised oncologic situation, because we had an emergent situation. Obviously, we would ideally diagnose these patients and initiate treatment, before they got to the point where they had cord compression.

Reference

  1. Jawad MU, Farhan S, Haffner MR, et al. Primary mobile vertebral column sarcomas: Prognostic factors vary by histologic subtypes. J Surg Oncol. Published online June 6, 2021. doi:10.1002/jso.26530