Clinical investigation: lung
Comorbidity and KPS are independent prognostic factors in stage I non-small-cell lung cancer

https://doi.org/10.1016/S0360-3016(01)02741-9Get rights and content

Abstract

Purpose: To determine the prognostic role of comorbidity in Stage I non-small-cell lung cancer (NSCLC) treated with surgery or radiotherapy (RT).

Methods and Materials: One hundred sixty-three patients with clinical Stage I NSCLC were analyzed for overall survival (OS) and comorbidity. One hundred thirteen patients underwent surgery (surgical group) and 50 patients received definitive radiotherapy (RT group). Ninety-six percent of the surgical group had lobectomy or pneumonectomy, and negative margins were achieved in 96% of the patients. The median dose to the tumor for the RT group was 61.2 Gy (range 30.8–77.4). The Cumulative Illness Rating Scale for Geriatrics (CIRS-G) and the Charlson scale were used to rate comorbidity. Karnofsky performance scores (KPS) were available in 42 patients; the rest of the scores were determined retrospectively by two physicians independently, with 97% agreement.

Results: The OS was 44% for the surgical group and 5% for the RT group at 5 years. Noncancer-related mortality was observed in 31% and 62% of the surgical and RT patients, respectively. On univariate analysis, performed on all patients (n = 163), squamous cell histologic type (p <0.001), clinical Stage T2 (p = 0.062), tumor size >4 cm (p = 0.065), >40 pack-year tobacco use (p <0.001), presence of a CIRS-G score of 4 (extremely severe, CIRS-G4: [+]) (p <0.001), severity index of >2 (p <0.001), Charlson score >2 (p = 0.004), KPS <70 (p <0.001), and treatment with RT (p <0.001) were associated with a statistically significant inferior OS. Multivariate analysis with histologic features, clinical T stage, age, tobacco use, KPS, comorbidity [CIRS-G(4)] and treatment group on all patients showed that squamous cell histology, >40 pack-year tobacco use, KPS <70, and presence of CIRS-G(4) were independently associated with an inferior OS. Treatment modality, T stage, and age did not have any statistically significant effect on OS. Statistically significant differences were found between the surgical and RT groups in Charlson score (p = 0.001), CIRS-G total score (p = 0.004), severity index (p = 0.006), CIRS-G4(+) (p <0.001), KPS (p <0.001), amount of tobacco use (p = 0.002), clinical tumor size (p <0.001), clinical T stage (p = 0.01), forced expiratory volume in 1 s (p = 0.001), and age (p = 0.008), in favor of the surgical group.

Conclusion: The presence of significant comorbidity and KPS of <70 are both important prognostic factors, but were found to be independent of each other in Stage I NSCLC. Therefore, comorbidity and KPS assessment are recommended when analyzing the prognostic effects of tumor or treatment-related factors on OS.

Introduction

Although the recommended treatment for Stage I non-small-cell lung cancer (NSCLC) is surgical resection, this approach is not possible in all patients. Comorbidity, functional status, pulmonary function, and age often determine the resectability in this group of patients. Unfortunately, the selection criteria of patients for surgical resection are not uniform. Therefore, the comparison of outcomes in different studies using the same treatment modality is difficult and can be misleading, because most of these studies lack any description of comorbidity and functional status.

Patients who are poor risk for surgery secondary to comorbid conditions, decreased pulmonary reserve, or functional status are often treated with definitive radiation. Despite having clinical stages similar to resected patients, these patients have a worse prognosis, limiting the ability to compare treatment modalities. Additionally, nonuniform surgical exclusion criteria limit the ability to compare radiation series, which are largely composed of inoperable patients. Noncancer-related deaths constitute varying percentages of all deaths in both surgical and radiotherapy (RT) series, reflecting differences in the patient populations in terms of functional and comorbidity status 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17 (see Table 3, Table 4).

In this retrospective study, we used two different comorbidity scales for patients treated with surgery or RT for Stage I NSCLC at the Medical College of Wisconsin (MCW) to evaluate the effects of comorbidity on overall survival (OS) and its prognostic importance.

Section snippets

Surgical group

The records of 113 consecutive patients with clinical Stage I NSCLC who had undergone surgery at MCW affiliated hospitals between June 1990 and November 1998 were reviewed (Table 1). Seventy-eight patients were treated at the Zablocki Veterans Affairs (VA) Hospital and 35 patients at a MCW affiliated community hospital. Most patients were men (87%). The median age was 66 years (range 46–80).

All patients were evaluated with a complete history and physical examination, chest X-ray, CT of the

Surgical group

Fifty-six percent (63 of 113) of clinical Stage I patients treated surgically were dead at the time of analysis. Twenty-eight of these 63 had died of disease recurrence and 35 patients of noncancer-related causes. The OS was 56% at 3 years and 44% at 5 years (Fig. 1). OS was not significantly different for patients with clinical or pathologic Stage T1 tumors compared with those with Stage T2. However, the presence of positive mediastinal or hilar lymph nodes on pathologic examination was

Discussion

Overall 5-year survival rates for surgical treatment of stage I NSCLC (pathologic) ranging from 43% to 84% have been reported in the literature (Table 3)3, 4, 5, 6, 7, 8, 9. Differences in tumor-related factors (T stage, tumor size, histologic features, grade) and surgical approach (pneumonectomy, lobectomy, segmentectomy, wedge resection) partially account for this wide range. It is also likely that patient selection based on comorbid conditions, performance status, age, and pulmonary function

Conclusions

The presence of significant comorbidity and KPS are important prognostic factors in Stage I NSCLC, independent of other factors. The evaluation of the effect of tumor or treatment-related prognostic factors on OS in the absence of comorbidity assessment or KPS may influence the accuracy of prognostic estimates and interpretation of results. Furthermore, it makes comparison of different studies difficult as patient selection for surgical resection is influenced by comorbidity and KPS. We

References (37)

  • B.J Slotman et al.

    Curative radiotherapy for technically operable stage I nonsmall cell lung cancer

    Int J Radiat Oncol Biol Phys

    (1994)
  • M.E Charlson et al.

    A new method of classifying prognostic comorbidity in longitidunal studiesDevelopment and validation

    J Chron Dis

    (1987)
  • M.D Miller et al.

    Rating chronic medical illness burden in geropsychiatric practice and researchApplication of the Cumulative Illness Scale

    Psychiatry Res

    (1992)
  • A.R Feinstein et al.

    Cancer of the larynx

    J Chronic Dis

    (1977)
  • J.D Clemens et al.

    A new clinical-anatomic staging system for evaluating prognosis and treatment of prostatic cancer

    J Chronic Dis

    (1986)
  • J.C Nesbitt et al.

    Survival in early-stage non-small cell lung cancer

    Ann Thorac Surg

    (1995)
  • P.A Kupelian et al.

    Prognostic factors in the treatment of node-negative nonsmall cell lung carcinoma with radiotherapy alone

    Int J Radiat Oncol Biol Phys

    (1996)
  • D.H Harpole et al.

    Stage I nonsmall cell lung cancerA multivariate analysis of treatment methods and patterns of recurrence

    Cancer

    (1995)
  • Cited by (0)

    1

    Dr. Bousamra’s current address is University of Louisville, Suite 1200, Rudd Heart and Lung Institute, 201 Abraham Flexner Way, Louisville, Ky 40201.

    View full text