Age-Adjusted Charlson Comorbidity Index in Recurrent Glioblastoma: A New Prognostic Factor?

Melanie Barz, Stefanie Bette, Insa Janssen, Amir Kaywan Aftahy, Thomas S. Huber, Friederike Liesche‐Starnecker, Yu‐Mi Ryang, Benedikt Wiestler, Stephanie E. Combs, Bernhard Meyer, Jens GemptResearch Square 2020 · DOI

Abstract

Background: For recurrent glioblastoma (GB) patients, several therapy options have been established over the last years such as more aggressive surgery, re-irradiation or chemotherapy. Age and the Karnofsky Performance Status Scale (KPSS) are used to make decisions for these patients as these are established as prognostic factors in the initial diagnosis of GB. This study’s aim was to evaluate preoperative patient comorbidities by using the age-adjusted Charlson Comorbidity Index (ACCI) as a prognostic factor for recurrent GB patients.

Methods: In this retrospective analysis, 133 patients who underwent surgery for primary recurrence of GB from January 2007 until December 2016 were included (49 females, 84 males, mean age 58 years). Preoperative age, sex, ACCI, KPSS, and adjuvant treatment regimes were recorded. Extent of resection (EOR) was recorded as a complete or incomplete resection of the contrast-enhancing tumor part.

Results: Median overall survival (OS) after initial diagnosis was 20.0 months (95% CI 17.2-22.9) and 9.0 months (95% CI 7.1-10.9) after first re-resection. Preoperative KPSS > 80% (P < 0.001) and ACCI < 7 (P = 0.020) were associated with significantly improved survival in univariate analysis. In multivariate analysis, preoperative KPSS (≥80/<80) was the only significant prognostic factor (HR 1.934, P = 0.003), while ACCI (<7/≥7) did not reach statistical significance (HR 1.608, P = 0.154).

Conclusion: ACCI may serve as an additional prognostic factor, particularly for patients with multiple comorbidities. However, it does not provide additional prognostic information beyond the well-established KPSS.

Introduction

Glioblastoma (GB) is the most aggressive primary brain malignancy. Despite standard-of-care initial treatment, recurrence is nearly universal. When a tumor returns, clinicians must decide if the potential survival benefit of a second surgery outweighs the procedural risks. Currently, these decisions rely heavily on chronological age and the Karnofsky Performance Status Scale (KPSS), a 0–100% score measuring a patient's functional independence.

As the patient population ages, many present with multiple comorbidities, such as diabetes or heart disease. This study evaluates whether the Age-Adjusted Charlson Comorbidity Index (ACCI)—a tool that weights 19 medical conditions alongside age—can offer a more nuanced and accurate prognosis than the standard KPSS in the setting of recurrent glioblastoma.

Methods and Patient Cohort

This retrospective, bicentric study analyzed 133 consecutive patients who underwent surgery for their first glioblastoma recurrence between 2007 and 2016. The cohort consisted of 49 females and 84 males with a mean age of 58 years. Patients whose tumors progressed from lower-grade gliomas or those on subsequent recurrences were excluded to maintain consistency.

The ACCI was calculated by scoring 19 conditions (weighted 1–6 points) with additional points added for every decade of life over 50. Surgical evaluation included tracking the Extent of Resection (EOR) to determine if complete removal of the contrast-enhancing tumor was achieved.

Table 1: Patient demographics and clinical characteristics of the study cohort (n=133).

Key Results

The median overall survival for the cohort was 20 months from initial diagnosis and 9 months following the first re-operation.

Statistical analysis revealed the following:

  • Univariate Analysis: Both an ACCI < 7 (P=0.020) and a KPSS ≥ 80% (P<0.001) were associated with significantly improved survival.
  • Multivariate Analysis: When factors were considered together, the KPSS (≥80%) emerged as the only significant preoperative prognostic factor (Hazard Ratio 1.934, P=0.003). The ACCI failed to maintain statistical significance (P=0.154).
  • Surgical Impact: The Extent of Resection (EOR) remained a powerful predictor; patients achieving complete resection of the recurrence had significantly better outcomes.

Discussion

The study's findings suggest that for brain tumor patients, functional status (KPSS) is a much stronger predictor of survival than a formal medical history (ACCI). This indicates that "what a patient can do" is more clinically relevant than "what medical conditions they have." For instance, a patient with controlled chronic illnesses but high mobility often fares better than a patient with fewer comorbidities but significant neurological deficits.

Furthermore, chronological age alone was not a significant predictor of survival after the first re-resection in this cohort. This suggests that biological fitness and tumor biology are more critical than a patient's birth date. The ACCI appears to be a "weak" prognosticator in this context because it overlaps significantly with the information already captured by the KPSS.

Conclusion and Limitations

While the ACCI correlates with survival in simpler models, it does not add significant prognostic value over the KPSS in multivariate assessments. The primary limitation of this study is its retrospective design, which may introduce selection bias. Additionally, while a sample of 133 is robust for this field, larger prospective studies are needed to determine if the ACCI provides benefits for specific high-comorbidity subgroups. Currently, the KPSS remains the primary tool for clinical assessment in recurrent glioblastoma.

Summary updated

Abstract

<title>Abstract</title> <bold><italic>Backround: </italic></bold>For recurrent glioblastoma (GB) patients, several therapy options have been established over the last years such as more aggressive surgery, re-irradiation or chemotherapy. Age and the Karnofsky Performance Status Scale (KPSS) are used to make decisions for these patients as these are established as prognostic factors in the initial diagnosis of GB. This study’s aim was to evaluate preoperative patient comorbidities by using the age-adjusted Charlson Comorbidity Index (ACCI) as a prognostic factor for recurrent GB patients.<bold><italic>Methods: </italic></bold>In this retrospective analysis we could include 133 patients with surgery for primary recurrence of GB from January 2007 until December 2016 (49 females, 84 males, mean age 58 years (range 21–80 years)). Preoperative age, sex, ACCI, KPSS and adjuvant treatment regimes were recorded for each patient. Extent of resection (EOR) was recorded as a complete/incomplete (including biopsy) resection of the contrast-enhancing tumor part.<bold><italic>Results: </italic></bold>Median overall survival (OS) after initial diagnosis was 20.0 months (95% confidence interval (CI) 17.2-22.9)) and 9.0 months (95% CI 7.1-10.9 months) after first re-resection. Preoperative KPSS &gt; 80% (P&lt;0.001) and ACCI &lt;7 (P=0.020) were associated with significantly improved survival in univariate analysis. Including these preoperative factors in multivariate analysis, preoperative KPSS (≥80/&lt;80) is the only significant prognostic factor (HR 1.934[1.3-3.0], P=0.003), whereas ACCI (&lt;7/≥7) missed statistical significance (HR 1.608[0.8-3.1], P=0.154). <bold><italic>Conclusion: </italic></bold>ACCI might be an additional prognostic factor for patients with recurrent glioblastoma, especially in those with many comorbidities. However, besides the well-established KPSS, the ACCI does not add further information about patients’ prognosis.
BibTeX
@inproceedings{barz2020age,
  author = {Melanie Barz and Stefanie Bette and Insa Janssen and Amir Kaywan Aftahy and Thomas S. Huber and Friederike Liesche‐Starnecker and Yu‐Mi Ryang and Benedikt Wiestler and Stephanie E. Combs and Bernhard Meyer and Jens Gempt},
  title = {{Age-Adjusted Charlson Comorbidity Index in Recurrent Glioblastoma: A New Prognostic Factor?}},
  booktitle = {Research Square},
  year = {2020},
  doi = {10.21203/rs.3.rs-113876/v1},
}
Figure