Thank you for that amazing introduction to this session. I have actually a lower rate, but now I'm going to have to update this slide. Twenty percent of cancer patients will develop spine metastases over the course of their illness. We've seen an increased number of metastatic spine tumors over the past decade, as MR and FDG-PET have improved detection, and these newer systemic treatments have improved survival. However, the biologic and checkpoint inhibitors are much more effective for visceral than for bone disease, so we think we're going to see a continued uptick in the number of spine tumors that we're responsible for. That being said, in 2018, the METEOR trial showed an improved progression for survival of renal cell to bone using cabozatinib compared to everolimus, but we think this is the exception and not the rule. The goals of treatment for metastatic disease are palliative. We want to improve or maintain neurologic function, achieve local tumor control, mechanical stability, pain relief, and ultimately improve quality of life. Far and away, the single greatest advance in the treatment of metastatic spine tumors is stereotactic radiosurgery. It's defined as high-dose per fraction conformal radiation, and the spine is typically given a 16- to 24-grade single fraction or 8- to 10-grade times 3. Compared to conventional radiation, often given as 3-grade times 10 fractions, radiosurgery has a shorter treatment time but can actually give a cytotoxic ablative tumoral dose, and that has fundamentally changed our treatment paradigms, including the indications for and type of surgery that we do, and that has led to this concept of separation surgery. There are major advances in radiosurgery over the past decade, including technological, defining tumor-cidal doses and dose-constrained organs at risk, and even defining the radiobiology of radiosurgery. The NOMES framework was developed in order to facilitate decision-making for metastatic spine tumors. There are four sentinel decision points, neurologic, oncologic, mechanical stability, and systemic disease, to decide best treatment. The neurologic and oncologic considerations are made in combination. From a neurologic perspective, we're very concerned about myelopathy and functional radiculopathy, but what we're principally concerned about is the degree of spinal cord compression, and a scoring system that has been validated to facilitate decision-making for these patients. The oncologic perspective is really how we control a tumor, and that is completely predicated on the radiation response. Radiation sensitivity has been completely redefined as we've transitioned from conventional radiation to stereotactic radiosurgery. We additionally now have brachytherapy with P32 plaques to clear dural margins, and high-dose rate brachytherapy with iridium afterloaded catheters to treat recurrent bone disease. Mechanical instability is a separate assessment, as no amount of radiation will stabilize an unstable spine. The SINS criteria was developed to define instability in the neoplastic setting, and once that's been determined, patients are often candidates for kyphovertebralplasty, subcutaneous pedicle screws, or open surgery. And as Zia discussed, the systemic disease and medical comorbidities are critically important in decision-making. Survival is a big determinant in this category, and with the new checkpoint inhibitors and biologics, survival has been extended for virtually every cancer. Here, a single agent in a volume of a PD-1 inhibitor has increased survival over three times compared to standard chemotherapy. There's very little in the literature that looks at conventional external beam radiation based on differential tumor histology, but what's there is remarkably consistent. There are radiosensitive tumors, such as the hematologic malignancies, breast and prostate cancer, versus the radio-resistant tumors, which are virtually the remainder of the solid tumors. For the radiosensitive tumors, we have two-year local control rates of 86% versus the radio-resistant tumors with two-year local control rates of only 30%. With stereotactic radiosurgery, we overcome radio resistance, and all tumors become radiosensitive. In our recent series, we have two-year local control rates of 98% that are histology independent. Here's a patient with multiple myeloma, a very radiosensitive tumor. It got three grade times 10 fractions, and you can see the tumor completely apoptosed, decompressing the spinal cord. The problem is you will never see these responses in the radio-resistant tumors, and for that reason, for patients with minimal or no spinal cord compression with radio-resistant tumors, we'll typically take them for upfront stereotactic radiosurgery. Here's our series of 811 patients, most of which were radio-resistant tumors at a median follow-up of over two years. Contours were according to radiosurgery consortium guidelines. The prescription dose was 18 to 26 gray single fraction, but dose was analyzed as a continuous variable, and an optimal cut point was set to establish a low versus high-dose group. The median dose of the planning target volume in the low-dose group was 16.4 gray versus the high-dose group, 22.4 gray. The only significant factor in the incidence of local failure was the dose of radiation. In the low-dose group, in a year, there was a 5 percent failure rate versus 0.4 in the high-dose group, and at four years, the low-dose group increased to 20 percent failure rate versus only 2.1 percent in the high-dose group. What is most important is that these responses are histology-independent. We overcome radio-resistance with high-dose perfraction radiation, and that has fundamentally changed some of our surgical indications. Here's a patient with a T10 solitary renal cell metastasis. If you use the Takahashi or Tamida scores, this would be recommended for an on-block resection. These are very long operations with massive blood loss, and the control rates are not well established. With radiosurgery, we have treatment times of 20 minutes, no blood loss, and 98 percent tumor control, and so there's a movement for patients with solitary metastases to go to radiosurgery rather than these massive on-block surgeries, resections. Dose constraints have been another fundamentally important determinant over the last 10 years. The most important constraint establishes for the spinal cord. At a cord maximum dose of 14 gray, we have less than a 1 percent chance of creating myelitis, but it's this dose constraint to the spinal cord that really prevents us from being able to treat high-grade cord compression. This is predicated in part on a number of pieces of evidence that we have in the literature. Mike Lovelock did a dose failure analysis where all local failures received less than 15 gray to some portion of the planning target volume, and if our cord D max is 14 gray with a 10 percent per millimeter fall-off, then either we get a cytotoxic tumor dose that risks overdose in the spinal cord, or we underdose at the margin of the cord and risk epidural progression. Secondly, the resolution of soft tissue disease in solid tumors can take months, unlike the myeloma I showed you, so we get no effect of immediate decompression of epidural disease. And finally, Sam Rue tried radiosurgery for high-grade cord compression, and despite a 48 percent loss to follow-up in that series, 20 percent of patients with radio-resistant disease demonstrated neurologic progression. Based on that, patients with high-grade cord compression with radio-resistant tumor are often recommended for upfront surgery followed by radiation therapy, and as Zia showed you, this is largely based on the PATCHEL study published in Lancet 2005. It was a prospective randomized trial looking at solid tumors, high-grade cord compression with myelopathy comparing surgery and conventional radiation to conventional radiation alone, and surgery was better in every outcome variable, and based on this and a large number of other series, a strong recommendation had been made that patients with high-grade spinal cord compression due to solid tumor malignancy undergo surgical decompression and stabilization followed by radiation therapy. The question became, what kind of surgery and what kind of radiation? With the integration of stereotactic radiosurgery, the neurologic goals of spinal cord decompression for neurologic salvage and mechanical stability using stereoscopic screw rod systems remains the same, but the oncology of how we get local tumor control, again, is completely predicated on the radiation response. When we used conventional external beam radiation, we were very aggressive and tried to do maximum cytoreductive surgery, either gross total resection or on block, with the expectation that conventional radiation wouldn't treat these tumors any better in the postoperative setting than it did in the upfront setting. With radiosurgery, our only goal really is to reconstitute the thecal sac and create a target for the radiation, and that became known as separation surgery. Here's an 86-year-old with papillary thyroid cancer with three-level vertebral body disease, and it's almost inconceivable that you could get this patient through a gross total or on block in order to get neurologic salvage and tumor control. And so many of these patients are now addressed through separation surgery. We usually will do a spinal fixation, two levels above and below the index level. We drill with a three-millimeter matchstick, the spinous process lamina, superior and inferior facet joint and pedicle flush the vertebral body. We then start normal dural margins to strip the tumor off the dura and then take out the posterior longitudinal ligament to affect a margin on the anterior dura. And so what you're ultimately trying to achieve, again, is reconstitution of the thecal sac to create a target for the radiation, which you can see on the myelogram for simulation purposes. We left all the vertebral body tumor behind. We did a long posterior fixation. Got eight grade times three, fully ambulatory, two years out with well-controlled disease. We gave her a surgery that she could tolerate and got control of the tumor with radiosurgery. What's the date on conventional radiation as a post-operative adjuvant? There's one series in the literature. Clay, Camp, Sammy published 1998. They did very aggressive resections on these patients with conventional external beam radiation as a post-op adjuvant. The local control was only 30 percent at one year. If you lived long enough, everybody recurred. And one of the biggest predictors of recurrence was tumor histology. Here's our series of separation surgery followed by radiosurgery in 186 patients. Most were operated for high-grade cord compression with radio-resistant tumor, and 50 percent had failed prior radiation therapy. There were three-dose strategies, high-dose single-fraction, high-dose hypofractionated, and low-dose hypofractionated. The one-year cumulative incidence of recurrence was 16 percent, but if we gave a high-enough-dose single- or high-dose hypofractionated radiation, we had less than 10 percent recurrence at a year. There were no neurologic complications associated with the radiation. And of note, again, these responses are tumor histology independent. Seventy-five percent of our non-ambulatory patients regained the ability to walk, and overall, 90 percent were ambulatory. One of the advantages of separation surgery over more aggressive approaches to these tumors, if you look at this tumor on the right, we have a multi-compartmental tumor with very high-grade cord compression. If we do a very aggressive approach to this, we'd have to do a front-back to take this tumor out and, again, not well-tolerated in the cancer population. With separation surgery, we have very short operating times of two-and-a-half hours. We don't need an approach surgery to go anterior transcavitory. And ultimately, we do a better and safer epidural decompression because we can always identify normal dura and strip the tumor off from known anatomy. In terms of the approach morbidity, here's the difference between a C1-2 approach for a primary tumor. This is a low-grade chondosarcoma, and the best approach for this is a transmandibular osteotomy to come through the jaw to get to this tumor in order to resect it. And the morbidity from that is significant. Patients need to be pegged and trached for two weeks after the surgery. And in the metastatic population, they're simply not going to tolerate that. That's in counter-distinction to this patient, a 33-year-old female who was pregnant, three months pregnant. She tripped coming off an airplane jetway and became acutely quadriplegic. She had this tumor identified and had an emergency decompression, but we didn't think we could go through the jaw to take out that tumor in a pregnant woman with high-grade core compression and was paralyzed. And so we simply did separation surgery, occipito-cervical fixation. By the grace of goodness, she recovered to normal neurologic function by two months, and the tumor ended up being leiomyosarcoma. She got IGRT and conventionally fractionated high-dose radiation. Here's the child who was born three months after the surgery. This is at the age of four, but you can look at the amount of tumor that we left behind. We got neurologic recovery and stabilized the spine, but we left that tumor behind. She got IGRT. Here's her 10-year follow-up film. It's not the amount of tumor you take out that controls it. It's ultimately the response to the radiation that defines whether they're going to get control of that tumor. One of the criticisms of separation surgery is that if we don't reconstruct the anterior column, these patients are going to fall apart. We looked at 318 patients, and the fixation failure rate was only 2.8%. In this patient population, it's a very robust reconstruction for their stability. Finally, we thought there was potentially an advantage to radiosurgery over conventional radiation in terms of wound complications because we can bring the beams in from multiple different trajectories as opposed to straight through the wound. We looked at patients who got preoperative radiation and ultimately went to surgery. With conventional radiation, we had a 17% risk of wound complications versus stereotactic radiosurgery, only a 6% wound complication rate. We think if we actually contour the beams intentionally to keep them out of the operative corridor, we can reduce that even further. The problem with that is if you get a wound complication, you can't simply do a washout. They actually need vascularized rotation flaps to fix these wounds. Although this is very successful, it adds significant morbidity to their treatment. In the NOMS approach, again, there's still a role for conventional radiation for radiosensitive tumors regardless of the degree of cord compression. For the radio-resistant tumors with minimal spinal cord compression, we're now going straight to stereotactic radiosurgery. For radio-resistant tumors with high-grade cord compression, the absolute workhorse of separation surgery reconstitute the thecal stack to create a target for the radiation and then let the radiation get control of the tumor. If the patient's unstable, they need a stabilization procedure, and everything is predicated on what the patient can tolerate from a systemic disease standpoint. Thank you.

metastatic spine tumors

cancer patients

spine metastases

stereotactic radiosurgery

treatment decisions