The impact of prior radiation or chemotherapy on surgical outcomes in the treatment of spine metastases

Abstract

Study Design:

This was a retrospective cohort study.

Objective:

The objective of this study was to evaluate the impact of prior chemotherapy and/or radiation on outcomes among patients undergoing surgery for spinal metastases.

Summary of Background Data:

Spinal metastases are associated with significant pain, functional decline, and poor prognosis. Prior chemotherapy and radiation are common and impose a substantial morbidity, but their influence on surgical outcomes remains poorly defined.

Materials and Methods:

A retrospective chart review of adult patients who underwent surgery for spine metastasis was conducted at a single institution (2017–2022). Demographic, surgical, and oncologic data were collected via manual chart review. Patients were stratified based on a history of chemotherapy and/or radiation. Outcomes included estimated blood loss, length of stay (LOS), rate and number of inpatient complications, 30-day emergency department visits, 90-day readmissions, 1-year reoperation, and 1- and 5-year mortality.

Results:

Forty-four patients were included (23 received prior chemotherapy/radiation). The most common surgical indication was multiple myeloma in patients with prior therapy (17.39%) and lung metastases in patients without prior therapy (28.57%). Seventeen patients (80.95%) without prior therapy received chemotherapy/radiation postoperatively. Patients with prior therapy experienced shorter LOS (9.78 ± 7.40 vs. 20.0 ± 16.0 days; P = 0.015), lower complication rate (39.1% vs. 76.2%; P = 0.030), and fewer complications per patient (0.74 ± 1.51 vs. 2.29 ± 1.82; P = 0.001). All other outcomes were similar between the groups.

Conclusions:

Prior therapy was associated with reduced LOS and inpatient morbidity. These findings may reflect differences in tumor biology or oncologic patient heterogeneity. Primary tumor types were poorly isolated, and the sample size limited definitive conclusions, further underscoring the need for robust future investigation.

INTRODUCTION

The spine is the most common site for metastatic spread to bone and accounts for over 70% of osseous metastases.[1] Due to advancements in medical technologies and treatments in recent decades, the cancer death rate has dropped substantially, and patients are experiencing longer life expectancies.[2] As a result, the overall prevalence of patients with cancer has increased, and in conjunction, the incidence of spinal metastases has also increased.[3] Approximately 10% of all cancer patients develop metastatic spinal lesions at some point in their disease course.[1] The clinical impact of spinal metastasis can be devastating. Most commonly, patients will experience severe pain, which limits their ability to engage in daily activities.[4] Less commonly, patients may experience motor and/or sensory dysfunction.[5]

The spectrum of treatment options to address spine tumors is broad. Surgical intervention for spinal tumors is generally indicated for severe pain, instability, or neurologic compromise refractory to other treatment attempts.[6] While surgery often may not represent a curative treatment, it can offer substantial palliative benefit by removing painful lesions, stabilizing the spinal column, and decompressing neural elements if needed. Surgical outcomes vary and depend on several patient-specific factors including tumor type, age, disease duration, and previous attempted treatment modalities, including chemotherapy and radiation.[7] The development of cancer-specific chemotherapy regimens has improved efficacy.[8] Radiation is another modality to consider, especially in cases with spine tumors that are less sensitive to chemotherapy. In recent decades, there has been an increase in the use of radiation treatment for spine tumors owing to advancements in radiation technology, including the advent of proton therapy, carbon ion therapy, intensity-modulated radiation therapy, and stereotactic body radiation therapy (SBRT).[9]

To date, there is limited research describing outcomes after surgery for spinal malignancies. A systematic review characterizing risk factors and preventive methods for complications after metastatic spine surgery found that only very low-quality evidence has been published regarding the relationship between neoadjuvant treatments and postsurgical infection or instrumentation failure.[10] Moreover, despite the increased use of chemotherapy and radiation prior to surgery for metastatic spinal disease, many key surgical outcomes including reoperation and mortality rates have not been well characterized. The present study aims to evaluate the effect of prior radiation and/or chemotherapy on surgical outcomes in patients undergoing surgery for cancer that has metastasized to the spine.

MATERIALS AND METHODS

Patient identification

A retrospective cohort study was initiated after obtaining approval from the Institutional Review Board. Adult patients (>18 years old) who underwent spine surgery at a single academic institution for management of metastatic disease to the spine during 2017–2022 were identified using a Structured Query Language search. To obtain the most comprehensive list of patients with spinal metastases, the current work queried for the International Classification of Diseases-10 codes C00-D49 among spine surgery patients. Manual chart review was used to confirm that the indication for spine surgery was spine metastasis.

Data collection

Upon identification of patients who fulfilled the inclusion criteria, manual chart review was used to supplement the initial data extraction and confirm demographic, surgical, and outcome data. Demographic parameters included age, sex, race, body mass index (BMI), smoking status, and Charlson Comorbidity Index (CCI). Preoperative oncologic data included primary tumor diagnosis, presence of spine metastases, and history of adjuvant therapy, which included chemotherapy and/or radiation to the spine. Postoperative adjuvant therapy data were also collected for patients who did not receive it preoperatively. Additional preoperative laboratory data included albumin, prealbumin, and transferrin levels to ascertain any nutritional differences between patients. Surgical outcome data included estimated blood loss (EBL), levels decompressed, length of stay (LOS), inpatient complications, 30-day emergency department (ED) visits, 90-day readmissions, 1-year reoperation, 1-year mortality, and 5-year mortality. Notably, 5-year mortality data were only obtained for patients in the 2017–2018 period.

Statistical analysis

Patients were stratified based on the presence or absence of a history of chemotherapy and/or radiation therapy prior to surgery for their spinal metastasis. Descriptive statistics were reported as mean and standard deviation for continuous variables, and number and percentage for categorical variables. A Shapiro–Wilk test was used to analyze the normality of continuous variables. Parametric data were analyzed using t-tests, while nonparametric data were analyzed using Mann–Whitney U tests. Categorical variables were compared using Pearson’s Chi-squared tests. Statistical significance was set to P < 0.05 for all analyses.

RESULTS

Forty-four patients were included who underwent spine surgery during 2017–2022 for spinal metastases (23 with prior chemotherapy and/or radiation). Patients with prior chemotherapy and/or radiation most commonly had surgery for multiple myeloma (17.39%), while patients without prior chemotherapy and/or radiation most commonly had surgery for lung metastases (28.57%) [Table 1].

Table 1.

Source of metastatic cancer stratified by history of chemotherapy and/or radiation

Cancer	No prior chemotherapy and/or radiation (n=21), n (%)	Prior chemotherapy and/or radiation (n=23), n (%)
Appendiceal	0	1 (4.35)
Breast	1 (4.76)	1 (4.35)
Cervical	0	1 (4.35)
Esophageal	0	2 (8.70)
Ewing sarcoma	0	1 (4.35)
Hepatocellular	2 (9.52)	0
Lung	6 (28.57)	3 (13.04)
Multiple myeloma	4 (19.05)	4 (17.39)
Pheochromocytoma	1 (4.76)	0
Prostate	1 (4.76)	3 (13.04)
Renal cell	4 (19.05)	3 (13.04)
Colon	1 (4.76)	1 (4.35)
Skin squamous cell	0	1 (4.35)
Lymphoma	0	1 (4.35)
Tonsillar squamous cell	0	1 (4.35)
Upper GI/pancreatobiliary adenocarcinoma	1 (4.76)	0

GI - Gastrointestinal

Impact of prior chemotherapy and/or radiation on outcomes

Among the patients who did not have prior therapy (chemotherapy and/or radiation), 17 (80.95%) received postoperative therapy on an outpatient basis following hospital discharge. Patients who received prior therapy were similar in terms of age, sex, race, BMI, CCI, smoking status, number of levels decompressed, and preoperative albumin, prealbumin, and transferrin compared to patients without prior therapy [Table 2].

Table 2.

Demographics stratified by history of chemotherapy and/or radiation

	No prior chemotherapy and/or radiation (n=21), n (%)	Prior chemotherapy and/or radiation (n=23), n (%)	P
Age (years)	65.7 (8.42)	61.0 (11.4)	0.128
Sex (female)	6 (28.6)	7 (30.4)	1.000
Race
White	16 (76.2)	18 (78.3)	1.000
Black	5 (23.8)	5 (21.7)
BMI (kg/m2)	28.7 (6.35)	28.0 (6.09)	0.707
CCI	6.52 (2.71)	6.83 (1.83)	0.895
Smoking status
Nonsmoker	6 (28.6)	10 (43.5)	0.506
Former smoker	9 (42.9)	9 (39.1)
Current smoker	6 (28.6)	4 (17.4)
Number of levels decompressed	2.90 (3.04)	2.48 (1.68)	0.879
History of chemotherapy	0	19 (82.6)
Current chemotherapy	0	6 (31.6)
History of radiation	0	16 (69.6)
Current radiation	0	3 (100)
Albumin	3.45 (0.68)	3.57 (0.61)	0.565
Prealbumin	20.2 (7.48)	20.4 (7.78)	0.935
Transferrin	208 (70.5)	192 (44.5)	0.622

BMI - Body mass index; CCI - Charlson comorbidity index

Patients who received prior therapy had shorter LOS (9.78 ± 7.40 vs. 20.0 ± 16.0 days; P = 0.015), a lower rate of inpatient complications (39.1% vs. 76.2%; P = 0.03), and a lower mean number of inpatient complications (0.74 ± 1.51 vs. 2.29 ± 1.82; P = 0.001) compared to patients without prior therapy. Patients who received prior therapy were otherwise similar in terms of EBL, 30-day ED visits, 90-day readmissions, 1-year reoperations, 1-year mortality, and 5-year mortality compared to patients who did not receive prior therapy [Table 3].

Table 3.

Surgical outcomes stratified by history of chemotherapy and/or radiation

	No prior chemotherapy and/or radiation (n=21), n (%)	Prior chemotherapy and/or radiation (n=23), n (%)	P
EBL (mL)	794 (661)	758 (653)	0.884
LOS (days)	20.0 (16.0)	9.78 (7.40)	0.015
Inpatient complication	16 (76.2)	9 (39.1)	0.030
Number of inpatient complications	2.29 (1.82)	0.74 (1.51)	0.001
30-day ED visit	4 (21.1)	4 (17.4)	1.000
90-day readmission	5 (26.3)	10 (45.5)	0.345
1-year reoperation	3 (15.8)	2 (10.0)	0.661
1-year mortality	3 (15.0)	8 (34.8)	0.257
5-year mortality	6 (35.3)	12 (57.1)	0.310

EBL - Estimated blood loss; LOS - length of stay; ED - Emergency department

The most common inpatient complication experienced by patients with (35.3%) and without (27.1%) prior therapy was related to cardiopulmonary issues. A breakdown of inpatient complications is provided in Table 4.

Table 4.

Inpatient complications stratified by history of chemotherapy and/or radiation

	No prior chemotherapy and/or radiation (n=21), n (%)	Prior chemotherapy and/or radiation (n=23), n (%)
Wound complications	1 (2.1)	0
Cardiopulmonary	13 (27.1)	6 (35.3)
Infectious	12 (25.0)	3 (17.6)
Neurologic	5 (10.4)	2 (11.8)
Endocrine	5 (10.4)	0
GI	4 (8.3)	0
Renal	6 (12.5)	4 (23.5)
Musculoskeletal	2 (4.2)	2 (11.8)

Complications by system: Wound complications - Wound infection; Cardiopulmonary - Acute respiratory distress syndrome/respiratory failure, atelectasis, laryngeal edema, pneumonia/COVID, new arrhythmia, pulmonary embolism, deep vein thrombosis, thrombophlebitis; Infectious - Pneumonia/COVID, urinary tract infection, bacteremia/sepsis; Neurologic - Altered mental status, seizure, posterior reversible encephalopathy syndrome, ataxia/decreased proprioception, Acute onset weakness; Endocrine - Symptom of inappropriate ADH, hyponatremia; GI - Small bowel obstruction, ileus; Renal - Acute kidney injury, urinary retention; Musculoskeletal - Persistent muscle spasms; GI - Gastrointestinal; COVID - Corona virus disease of 2019; ADH - Antidiuretic hormone

DISCUSSION

Cancer that metastasizes to the spine can cause significant disability for patients.[1] Surgery can offer meaningful relief from symptoms of pain, radiculopathy, and neurologic compromise.[11] However, patients with cancer represent a high-risk surgical population, and the decision to pursue operative management requires a consideration of the risks and benefits of such an approach. Thus, it is important to develop an understanding of the potential factors that influence outcomes of spine surgery in these patients. The present study evaluated the impact of a history of prior chemotherapy and/or radiation on surgical outcomes in patients undergoing surgery for osseous metastases of the spine.

In the current work, the distribution of cancer types was different between patients who did versus did not have a history of prior therapy. Patients with prior chemotherapy and/or radiation most commonly had surgery for multiple myeloma, and patients without prior chemotherapy and/or radiation most commonly had surgery for lung metastases. Otherwise, the two groups were demographically similar and underwent a similar number of levels decompressed. Notably, among patients who did not receive prior chemotherapy/radiation, 17 (80.95%) received chemotherapy/radiation following hospital discharge. Patients with prior therapy experienced shorter LOS and a lower rate and average number of inpatient complications. However, there were no differences in EBL, 30-day ED visits, 90-day readmissions, 1-year reoperations, or 1- or 5-year mortality between the groups.

The current work noted differences between the two groups in the primary cancer types that were metastatic to the spine. Additionally, although the two groups were stratified based on a history of prior chemotherapy and/or radiation, it is important to note that over 80% of the group without prior therapy received outpatient chemotherapy and/or radiation after discharge. These two factors emphasize the complexity inherent to oncologic spine surgery patients. The course and timing of nonoperative treatments for metastatic disease, including chemotherapy and radiation, are highly individualized for each cancer type and patient. Consequently, although the current work identified differences in short-term outcomes that appeared to favor the prior therapy group, it is not possible to draw strong conclusions based on this factor alone due to heterogeneity between the groups. Thus, spine surgeons performing preoperative planning, patient counseling, and risk stratification for spinal metastases should work closely with the patient’s oncology and any other necessary medical and surgical teams. The absence of prior chemotherapy and/or radiation should not be viewed in isolation as a positive prognostic factor without considering a patient’s entire oncologic history.

While effective, radiation therapy poses a significant risk of toxicity to the spinal cord and surrounding bone and can potentially impact surgical outcomes.[12] The effectiveness of radiation therapy stems from its ability to damage cancer cell DNA.[13] Historically, external beam radiation therapy was the most common approach to spine radiation; however, it was unable to provide targeted radiation, giving it a high potential for toxicity to surrounding native tissue.[12] The emergence of SBRT allows for a more targeted approach to radiation and limits overall toxicity.[12,14] Nevertheless, the proximity of important spinal structures still carries an associated risk. Clinical manifestations of radiation toxicity include acute symptoms such as radiation-induced myelopathy.[13] Chronic signs of toxicity can range from minor neurologic deficits to more severe deficits depending on the degree of damage to the spinal cord.[13] Furthermore, patients who receive spine radiation are also at increased risk of vertebral compression fractures due to radiation’s directly antiangiogenic effects on bone.[13,15] In a mouse model, Okunieff et al. showed that hind limb radiation reduced bone growth in a dose-dependent manner.[15] Taken together, these findings provide evidence that spine radiation may have significant implications for surgery outcomes in patients with spine metastases.[16,17]

We observed that patients with a history of radiation and/or chemotherapy had a lower rate and average number of inpatient complications and shorter LOS. This is interesting considering the aforementioned morbidity and risk associated with radiation therapy. There are a number of factors that might have led to this observation. First, due to limited sample size, it was necessary to analyze patients who had chemotherapy and/or radiation in aggregate, rather than in individual groups of those with chemotherapy or radiation. This likely contributed to study heterogeneity. Additionally, the distribution of cancer types was different between the groups. In particular, there was a higher rate of lung cancer metastases in the group without a history of therapy prior to surgery. Lung cancer is the most significant source of cancer mortality in the world, with a yearly rate of new diagnoses that remains high.[18,19] Lung cancer represents a particularly poor prognosis compared to other solid tumors that tend to metastasize to bone, such as breast and prostate cancer.[20,21,22,23,24] Thus, differences in cancer type may have portended a different risk profile between the two groups, regardless of the presence of prior therapy. Finally, although it could be interpreted that the group with prior therapy had more severe cancer status necessitating systemic therapy, a substantial percentage of the no prior therapy group had outpatient chemotherapy and/or radiation after discharge. Thus, the two groups were not necessarily at different stages based on the timing of systemic therapy.

For these reasons, it is difficult to comprehensively interpret the observation that patients who received systemic therapy prior to surgery actually experienced decreased length of hospital stay and burden of short-term complications. Due to the adverse effects of radiation, we initially hypothesized that patients may be at increased risk of wound complications and need for reoperation. While radiation is effective in damaging cancer cells, it may also have detrimental effects on nearby healthy tissue. Radiation depletes fibroblasts and impairs their function, resulting in a predisposition to necrosis.[25,26] Radiation has also been associated with increased susceptibility to wound infection due to a suppressed immune state.[27] Jarvers et al. showed that patients who underwent radiation within 21 days of surgery for spinal metastases were more likely to have a wound-related revision surgery.[25] On the contrary, in our cohort, patients who received prior radiation and/or chemotherapy were no more likely to experience an inpatient complication on bivariate analysis. Additionally, after the inpatient period, all outcomes equalized between the two groups. This could be secondary to the high rate of postoperative systemic therapy after discharge among the group who did not receive prior therapy.

Similar to radiation, the morbidity of chemotherapy regimens is also well established. Chemotherapy treatments generally act by inhibiting metabolism in cancer cells, causing an arrest of tumor development.[28] Due to the systemic effects of chemotherapy, rapidly replicating noncancerous cells are also often affected. As a result, patients receiving chemotherapy experience widespread immunosuppression and are at an increased risk of infection and other complications.[29,30,31] Nevertheless, chemotherapy is often a first-line treatment for most cancers and may be unavoidable. In our study, we found that patients who received prior chemotherapy and/or radiation had a lower rate of inpatient complications. This is contrary to what we initially hypothesized. In addition to the aforementioned differences in tumor type, which may have impacted inpatient complication rates, it is possible that patients who received chemotherapy prior to surgery may have had the benefit of reduced tumor burden. However, due to the limitations and heterogeneity inherent to the current dataset, it is not possible to investigate this possibility with more granularity. Further work with a greater number of patients is needed to examine these findings in order to perform subgroup analysis by cancer tissue type.

The current work has several limitations that are important to elucidate. First, due to its retrospective nature, there were several variables that we were unable to account for in our analysis, as previously mentioned. For instance, the present investigation did not account for the different types of radiation therapy or chemotherapy that patients received. Thus, the group who received prior therapy may have received a less morbid regimen compared to the large percentage of patients who received postdischarge therapy. Additionally, more granular oncologic details were not available, which also may have affected outcomes. Further work should include a greater number of patients and more details related to the intricacies of patients’ cancer treatment history. Third, due to the small size of the study cohort, it was not possible to perform subgroup analysis by cancer tissue type. As previously mentioned, the prior therapy and no prior therapy groups differed in the distribution of tissue types, which likely impacted results. Finally, 5-year mortality data were not available for all years evaluated due to limitations in the electronic medical record. Thus, the ability of the current work to evaluate long-term impacts from chemotherapy and/or radiation was limited. Despite these limitations, the current work provides a descriptive analysis of differences in surgical outcomes among patients undergoing surgery for spinal metastases stratified by the timing of systemic therapy.

CONCLUSIONS

The present study is among the first to investigate the impact of the timing of receiving chemotherapy and/or radiation relative to surgery for spinal metastases. There was a different distribution of cancer tissue types between the group that received therapy before surgery and the group that did not. Interestingly, the group that received prior therapy had a lower rate and average number of inpatient complications and a shorter LOS. This is potentially due to a higher rate of lung cancer metastases among patients who did not receive prior therapy. Despite these differences in inpatient outcomes, long-term outcomes such as ED visits, readmissions, reoperation, and mortality were similar between the groups. This observation is potentially due to a high rate of postdischarge chemotherapy and/or radiation among the no prior therapy group. In sum, these findings highlight the complexity and heterogeneity inherent both to research on spinal metastases and to the surgical and medical management of these tumors. Surgeons treating spinal metastases should work closely with the oncology team and other multispecialty teams involved in the care of these complex patients. Further work is needed with a greater number of patients to clarify the effect of chemotherapy and/or radiation on spine surgery outcomes in a tissue-specific manner.

Conflicts of interest

Mitchell Ng: Stryker. Inc. – paid consultant, Johnson and Johnson Ethicon Inc. – paid consultant, CurvaFix Inc. – paid consultant, Pacira BioSciences Inc. – paid consultant, Sage Products Inc. – paid consultant, Alafair Biosciences Inc. – paid consultant, Next Science LLC – paid consultant, Bonutti Technologies Inc. – paid consultant, Hippocrates Opportunities Fund LLC – paid consultant, and Ferghana Partners Inc. – paid consultant. Jose A. Canseco: Accelus – research support, Cervical Spine Research Society – board or committee member, PathKeeper Surgical – stock or stock options; unpaid consultant, Wolters Kluwer Health – Lippincott Williams and Wilkins – editorial or governing board. Alan S. Hilibrand: Biomet – IP royalties, CTL Amedica – IP royalties, Paradigm Spine – stock or stock options. Alexander R. Vaccaro: Receives royalties from Stryker, Globus, Medtronic, Atlas Spine, Alphatec Spine, Spine Wave, Spinal Elements, Curiteva, Elsevier, Jaypee, Stout Medical, Taylor Francis/Hodder and Stoughton, Wolters Kluwer, Wheel House Medical, and Thieme; has stock or stock options in Accelus, Advanced Spinal Intellectual Properties, Atlas, Avaz Surgical, AVKN Patient Driven Care, Cytonics, Deep Health, Dimension Orthotics LLC, Electrocore, Flagship Surgical, FlowPharma, Rothman Institute and Related Properties, Globus, Harvard MedTech, Innovative Surgical Design, Jushi (Haywood), Orthobullets, Parvizi Surgical Innovation, Progressive Spinal Technologies, Sentryx, Stout Medical, See All AI, and ViewFi Health; is a consultant for Curiteva, Medcura, Stryker, Globus, Spinal Elements, Accelus, Wheel House Medical, and Ferring Pharmaceutical; serves on Scientific Advisory Board/Board of Directors/Committee for National Spine Health Foundation (NSHF), Sentryx, and Accelus; and is a member in good standing/independent contractor for AO Spine. Gregory D. Schroeder: Advance Medical – paid consultant, Bioventus – paid consultant, Surgalign – paid consultant, AOSpine – board or committee member, AOSpine – other financial or material support, Cervical Spine Research Society – board or committee member, Cerapedics – research support, DePuy, A Johnson and Johnson Company – research support, Medtronic Sofamor Danek – research support, Wolters Kluwer Health – Lippincott Williams and Wilkins – editorial or governing board. Christopher K. Kepler: Clinical Spine Surgery – editorial or governing board, Inion – IP royalties, Curiteva - IP royalties, Regeneration Technologies, Inc. – research support, NIH - grant, CSRS - research support, Altus – equity. All other authors: None.

Funding Statement

Nil.