Effects of weekly paclitaxel or paclitaxel plus carboplatin on functionality and symptoms of geriatric patients with cancer as measured by a brief geriatric oncology module

A pilot experience


  • Presented, in part, to the American Society of Clinical Oncology, Chicago, IL, June 9, 2003.



To measure functionality, symptoms, and quality of life in elderly patients, a geriatric oncology module (GOM) was developed that could be rapidly and effectively administered in community oncology practices.


The GOM was validated by semi structured patient interviews. The authors tested 26 patients receiving chemotherapy with paclitaxel or paclitaxel plus carboplatin to determine the effect of drug treatment on functionality, symptoms, and quality of life (QOL).


Chemotherapy stabilized or improved functionality, symptoms, and QOL in 75%, 86%, and 76% of patients, respectively. This GOM was easily implemented in a busy community oncology practice.


Weekly paclitaxel and paclitaxel plus carboplatin produced measurable palliation in geriatric oncology patients. This GOM may be useful in the assessment of specific chemotherapeutic interventions in elderly patients with cancer in community oncology practices. Cancer 2005. © 2005 American Cancer Society.

Older patients with cancer have more comorbidities and are more sensitive to loss of functionality and development of serious symptoms and reduced quality of life (QOL). This pattern of increased symptoms and diseases often leads to shortened survival times as well as to a reduced entry of such patients into clinical trials.1, 2

As a result, increased attention has been paid in recent years to the special needs and special care required by geriatric oncology patients.3 Although certain well funded geriatric oncology research programs have procedures and staff specifically trained to care for and study geriatric patients, most oncology centers, either at universities or in communities, lack such resources, staffing, experience, skills, and research-oriented modules for monitoring such patients. However, older patients with cancer may benefit from these evaluations, and thus there is a need for the development of a geriatric oncology module (GOM) that can be applied in busy oncology programs.

In addition, geriatric patients may benefit from different cancer therapy regimens than younger patients with cancer. Because of this, different treatment regimens designed specifically for use in older patients have been developed for common cancers.4, 5 The benefits of such geriatric cancer therapy regimens might be characterized by a simple geriatric oncology evaluation methodology.

This pilot study was performed to evaluate two hypotheses. First, functionality and symptoms of geriatric patients can be measured efficiently in a busy community oncology practice using a simplified methodology. Second, weekly chemotherapy programs, using either paclitaxel alone or paclitaxel plus carboplatin, can preserve or improve functionality and symptoms in geriatric oncology patients.

We therefore developed an oncology module that could be applied quickly in a community oncology practice, and tested this module with a weekly chemotherapy program in geriatric patients.


Patient Eligibility

Patients were eligible for this program if they were ≥ 65 years. Patients had to have measurable recurrent or metastatic cancer. They also had to have satisfactory pretreatment, hematologic, hepatic, and renal function as measured by a leukocyte count > 3000 cells/mm3, a platelet count > 75,000 cells/mm3, a bilirubin level < 2.0 mg/dL, and a creatinine level < 2.0 mg/dL. Patients must have been able to complete the questionnaire and data forms on the geriatric module, either alone or with an interpreter. Voluntary informed consent was obtained from the patients before they were able to participate and receive paclitaxel or paclitaxel plus carboplatin. Three additional patients also received other chemotherapy regimens, but these patients were only used in the development of the GOM.

Patient Treatment

The patients received the following chemotherapy regimens. Single-agent paclitaxel 80 mg/m2 was given intravenously (i.v.) weekly for 3 weeks, with cycles repeated every 4 weeks. Paclitaxel plus carboplatin was given utilizing paclitaxel 80 mg/m2 i.v. weekly for 3 weeks plus carboplatin i.v. weekly for 3 weeks at a calculated area under the curve of 2. Treatment cycles were repeated every 4 weeks.

In addition, other chemotherapy regimens were used in the development of the GOM. One patient received paclitaxel and carboplatin on a schedule of every 3 weeks, 1 patient received pamidronate and goserelin every 4 weeks, and 1 patient received cisplatin every 4 weeks with gemcitabine weekly for 3 weeks, repeated every 4 weeks.

For all regimens used, dose adjustments and/or delays were implemented for hematologic and nonhematologic toxicities according to standard practice.

Antitumor Response Measurement

A partial response (PR) was defined as a reduction in the sum of the products of the cross-sectional diameters of tumor by > 50%. This could not be associated with progression of disease in any other location. Patients with prostate carcinoma were defined as having a PR in the absence of measurable tumor if their prostate-specific antigen level had decreased > 50%. Tumor progression was defined as a 50% increase in the sum of the products of the cross-sectional diameters of bidimensional tumor measurements.

Geriatric Oncology Module

The GOM consisted of five elements. A five-point pain scale, a five-point energy scale, and a five-point global QOL evaluation were developed (based on our previous experience in studies of QOL).6 These 3 scales are shown in Figure 1. The global five-point QOL scale was a modification of the global scale previously developed.6–9 Before its use in the current study, the scale was validated by a semi-structured interview.

Figure 1.

Scales for pain, energy, and quality of life.

In addition, the patients completed the activities of daily living scale (ADL) and the instrumental activities of daily living scale (IADL) as indicated in Figures 2 and 3, respectively.

Figure 2.

Activities of daily living scale.

Figure 3.

Simplified instrumental activities of daily living scale.

Evaluation of Response in Geriatric Oncology Module Scales

Patients were defined as responders in the GOM scores if there was an improvement or stability at any level for ≥ 1 month duration. In addition, alternate response criteria were evaluated as shown below in the Results section.

Validation of Module

Over a 12-month period, 300 patients completed the pain, energy, and QOL scale and then each patient participated in a semi-structured interview with the physician. The physician asked the patient and family members about pain, energy, and global QOL. Results were compared with scale records. There was a > 90% consistency rate, validating the accuracy of the scales. The ADL and IADL scales were standard and had been previously validated.


Patient Characteristics

Twenty-six patients were evaluated in the pilot study. Their median age was 75 years (range, 65–87 years). There were slightly more males than females (Table 1).

Table 1. Patient Characteristics (n = 26)
CharacteristicsNo. of patients
  1. NSCLC: nonsmall cell lung carcinoma.

Median age (range)75 (65–87 yrs)
 Adenocarcinoma of unknown primary1
 Paclitaxel weekly14
 Paclitaxel and carboplatin weekly9
 Other therapy3

A wide range of tumors was studied, but the most common tumors were nonsmall cell lung carcinoma (n = 8), prostate carcinoma (n = 7), and breast carcinoma (n = 3).

As listed in Table 1, a number of other tumors were also studied. This range of cancer types is common in elderly patients with cancer. Fourteen of the patients received paclitaxel, nine patients received carboplatin plus paclitaxel on a weekly schedule, and three patients received other therapy as indicated in the Materials and Methods section.

Pretreatment Scores

As seen in Table 2, the patients received mean scores that were somewhat below normal. On the ADL scale, based on a possible score of 26, the mean score was only 22 and for the IADL scale, based on a possible score of 24, the mean score was only 18.3. The average for patients indicated mild to moderate pain, moderate loss of energy, and mild to moderate reduction in QOL.

Table 2. Pretreatment Scores
CharacteristicsEvaluable patientsMean scoreNormal scorePossible range
  1. ADL: activities of daily living; IADL: instrumental activities of daily living; QOL: quality of life.


Changes in Functionality and Symptoms while Receiving Chemotherapy

Patients were monitored with the ADL and IADL scales during the course of their weekly chemotherapy with paclitaxel or paclitaxel plus carboplatin. As indicated in Table 3, 7 of 12 evaluable patients showed an improvement or no change on the ADL scale, despite receiving weekly chemotherapy. This indicated that, generally, patients did not show a deterioration while receiving the weekly chemotherapy.

Table 3. Functionality Changes while Receiving Weekly Chemotherapy ADL and IADL
CharacteristicsImprovedNo changeWorse
  1. ADL: activities of daily living; IADL: instrumental activities of daily living.


The changes in symptoms while receiving weekly chemotherapy are listed in Table 4. Improvement or stability in pain was reported by 15 of 19 patients. Of 22 patients, 19 showed improvement or stability in energy level, and 16 of 21 showed improvement or stability in global QOL. This also indicated that the effects of weekly chemotherapy as administered in this program did not result in substantial deterioration of energy level or QOL.

Table 4. Symptom Changes while Receiving Weekly Chemotherapy: Pain, Energy Level, QOL
CharacteristicsImprovementNo changeWorse
  1. QOL: quality of life.


The overall net effect of weekly chemotherapy on either functions or symptoms is summarized in Table 5. This indicated 58–86% stability or improvement in the 5 scores in the GOM in this pilot evaluation.

Table 5. Net Effect of Weekly Chemotherapy on Function and Symptoms: ADL, IADL, Pain, Energy Level, and QOL
CharacteristicsNo. stable or improved (%)
  1. ADL: activities of daily living; IADL: instrumental activities of daily living; QOL: quality of life.

ADL7/12 (58)
IADL9/12 (75)
Pain15/19 (79)
Energy19/22 (86)
QOL16/21 (76)

Responses in function and symptoms were then correlated with antitumor response. Of the patients, 17 were evaluable for antitumor response (Table 6), and 13 of 17 patients had a PR (76%). Ninety-two percent of patients who achieved a PR had a stable or improved score. However, 3 of 4 (75%) patients who showed no change in tumor or a worsening in tumor size also had stable or improved scores.

Table 6. Relationship of Antitumor Response to Function and Symptom Score
Antitumor responseNo. of patientsStable or improved score (%)
Partial response1312 (92)
No change22 (100)
Worse21 (50)

We also compared the ages of responders and nonresponders. The median age of pain scale responders was 74.2 years, whereas the median age of the nonresponders was 75.0 years; energy scale responders 75.8 years, nonresponders 77.7 years; QOL responders 79.8 years, nonresponders 75.8 years; ADL responders 76.7 years, nonresponders 76.4 years; and IADL responders 75.0 years, nonresponders 76.3 years. Thus, there was no clear correlation between symptom or function response and age.

Sensitivity Analysis of Response Criteria

Because response criteria vary, we performed a sensitivity analysis to determine the frequency of response according to definition of response. As analyzed in Table 7, the response rates varied considerably depending on the definition employed. The results described above used a definition of response as stable at any level or improved for ≥ 1 month. If such stability or improvement were required for > 2 months, results were slightly lower. Furthermore, when the definition is defined as stable at only a good level of functionality or symptoms or when the definition is defined as improvement to a good level of functionality or symptoms, results were less. Finally, if one accepted the definition of response as only an improvement in scores, suggesting that patients with a normal score could not demonstrate any improvement, then responses were much less frequent.

Table 7. Response Criteria Sensitivity Analysis
Responding category Definition
No. of patientsImproved onlyStable at a good level or improvedStable at any level or improved
>1 Mo>2 Mo> 1 Mo> 2 Mo> 1 Mo> 2 Mo
  1. ADL: activities of daily living; IADL: instrumental activities of daily living; QOL: quality of life.


Correlation with Survival

The overall median survival of all patients was 23.5 months. Although the sample was very small overall and the tumor diagnoses were diverse, we tried to compare survival and symptom/function responses using the GOM. Survival of patients with improved functionality or symptoms at landmarks of either 2 or 3 months was compared with the survival of patients with worsening of functionality or symptoms at the same time points. No differences were apparent, but there were few patients in each group.


The current study was performed to determine whether a simple GOM could be employed in a busy community oncology setting. The GOM was developed based on our previous experience with the development of QOL studies.6–9

The QOL, fatigue, and pain scores were validated over the course of 12 months of development using semi-structured interview processes at the California Cancer Medical Center (Los Angeles, CA). These interviews performed by different physicians demonstrated a correlation of scores with patient concerns and complaints. The rate of completion of such scores is generally 85%.

The ADL and IADL scales are now accepted tests for evaluating the functionality of geriatric patients. These scores were used without further modification. Patients were asked to complete the QOL, energy level, pain, and the ADL and IADL scales as well as a symptom inventory before being evaluated by the physician. The average patient completed these in 10–15 minutes. Physicians evaluated the scores in the course of a usual physician visit. This did not prolong the visit time.

The use of QOL scales, fatigue scales, pain scales, and the ADL and IADL scales is not conventional in oncology practices. Therefore, the current trial represented a pilot experience with the use of these modules to determine the relevance to oncology practice. This module was easy to administer and required only a short time for completion. Furthermore, this could be completed by the patient or the patient plus the family without additional help from medical assistants, nurses, or physicians. Scoring of the ADL and IADL scales was performed by medical assistants with only 15 minutes of training.

In the semi-structured interviews, patients said that the scales were easy to complete and were useful in communicating their symptoms to physicians. This is consistent with published data indicating that QOL assessment improves patient well-being.10

However, some of the scales were completed incorrectly and were not evaluable, despite patient feedback that the scales were “easy.” Our experience is such that elderly patients often fail to complete all scales when long instruments are used serially. This pilot study attempted to develop a short instrument that would have high compliance. The rates of evaluable responses were as follows: pain, 83%; energy, 96%; QOL, 91%; and the longer ADL and IADL forms, 52%.

Use of this module produced helpful information, which focused the physician on the complaints of the patient. For example, deterioration of pain scores prompted appropriate pain evaluations and interventions. Reduced energy scores prompted evaluation for anemia and use of erythropoietin or addition of other fatigue management intervention. Reductions in QOL not associated with other symptoms prompted evaluation of psychosocial concerns, sleep disorders, depression, and/or anxiety. Reductions in ADL or IADL prompted discussion of additional support mechanisms such as physical or occupational therapy necessary at the patients' home to improve functionality.

In addition, the GOM helped to evaluate the impact of weekly chemotherapy utilizing either paclitaxel or paclitaxel plus carboplatin. The results indicate that the use of these two chemotherapy regimens on a weekly basis preserved or improved functionality and symptoms in the majority of geriatric oncology patients for at least a short period of time. Therefore, weekly paclitaxel and paclitaxel plus carboplatin are effective palliative regimens in geriatric oncology patients.

Because there are no gold standards for symptoms and function response, we believed that it was necessary to perform a sensitivity analysis according to different criteria of response (Table 7). Once investigators can agree on standard criteria for symptom and function response, such as response evaluation criteria in solid tumors that is used for tumor response, future studies will not need to perform this analysis.

Because this was a pilot investigation, additional studies are needed. This GOM should be compared with more comprehensive, time-consuming, and validated measurements of QOL, energy, and pain as well as ADL and IADL results. Specifically, the QOL five-point scale should be compared with our standard comprehensive and longer QOL instrument.6–9 Furthermore, such evaluations should be performed in multiple sites to confirm the applicability of this module and its correlation with more intensive monitoring methods.

In summary, we recommend the use of this module in practices as an aid to the management of geriatric oncology patients. This module may be useful in the assessment of specific chemotherapeutic intervention in elderly patients with cancer.